after the blood is shed. The nucleo-albumin part is derived faom the corpuscles of the blood (leucocytes, blood-plates), which break down and go into solution. This nucleo-albumin then unites with the calcium salts present in the blood to form fibrin ferment, an organic compound of calcium capable of reacting with fibrinogen. The theory is a simple one; it accounts for the importance of calcium salts in coagulation, and reduces the interchange between fibrinogen and fibrin ferment to the nature of an ordinary chemical reaction; but it cannot be accepted without reservation at present, since the experimental evidence is not entirely in its favor. Hammarsten, for instance, in some careful experiments seems to have obtained facts that are at variance with a part at least of this theory. Hammarsten' states that blood-plasma or fibrinogen solutions to which an excess of potsssium oxalate had been added, and which therefore was free presumably from precipitable calcium salts, underwent typical coagulation when mixed with blood-serum to which an excess of oxalate had also been added. In other words, a solution of fibrinogen free from calcium reacted with a solution of fibrin ferment (blood-serum) also apparently free from calcium. It might be urged against this experiment, however, that in the blood-serum used the combination of calcium and nucleoproteid to form ferment had already taken place, and that in this combination the calcium is not acted upon by the oxalate. Hammarsten indeed admits that something of this kind may occur, for he is convinced, like others, that calcium in some way is essential to coagulation, his suggestion being that it plays an unknown part in the formation of the ferment. He supposes that in the plasma of shed blood a material is present which he designates as prothrombin, and the calcium in some way converts this into the active ferment, the thrombin. According to the more explicit hypothesis of Pekelharing, the prothrombin is a form of nucleo-proteid and the thrombin a calcium compound of this proteid. The second part of Pekelharing's theory, namely, that the reaction between the ferment and the fibrinogen consists in a transfer of the calcium from the former to the latter, is directly contradicted by Hammarsten's experiments. Quantitative analysis of fibrinogen and fibrin showed that the latter does not contain any larger amount of calcium than the former. This author is inclined to consider the Ca contained in fibrin of the nature of an impurity, and not as an essential constituent of the fibrin molecule. By the use of special methods he has succeeded in obtaining typical fibrin containing as little as 0.005 per cent. of Ca. We must be content to say that in the clotting of blood three factors are necessary-namely, the fibrinogen and the calcium salts of plasma, which are present in the circulating blood, and the fibrin ferment, which is formed after the blood is shed.

Nature and Origin of Fibrin Ferment (Thrombin).—Recent views as to the nature of fibrin ferment have been referred to incidentally in the description of the theories of coagulation just given. The relation of these newer views to the older ideas can be presented most easily by giving a brief description of the development of our knowledge concerning this body. 1 Zeitschrift für physiologische Chemie, Bd. 22, S. 333, and 1899, Bd. 28, S. 98.

1

Schmidt prepared solutions of fibrin ferment originally by adding a large excess of alcohol to blood-serum and allowing the proteids thus precipitated to stand under strong alcohol for a long time until they were thoroughly coagulated and rendered nearly insoluble in water. At the end of the proper period the coagulated proteids were extracted with water, and there was obtained a solution which contained only small quantities of proteid. It was found that solutions prepared in this way had a marked effect in inducing coagulation when added to liquids, such as hydrocele liquid, that contained fibrinogen, but did not clot spontaneously or else clotted very slowly. It was afterward shown that similar solutions of fibrin ferment are capable of setting up coagulation very readily in so-called salted plasma-that is, in blood-plasma prevented from clotting by the addition of a certain quantity of neutral salts. It was not possible to say whether the coagulating power of these solutions was due to the small traces of proteid contained in them, or whether the proteid was merely an impurity. The general belief for a time, however, was that the proteids present were not the active agent, and that there was in solution something of an unknown chemical nature which acted upon the fibrinogen after the manner of unorganized ferments. This belief was founded mainly upon three facts: first, that the substance seemed to be able to act powerfully upon fibrinogen, although present in such minute quantities that it could not be isolated satisfactorily; second, it was destroyed by heating its solutions for a few minutes at 60° C.; and, third, it did not seem to be destroyed in the reaction of coagulation which it set up, since it was always present in the serum squeezed out of the clot. Schmidt proved that fibrin ferment could not be obtained from blood by the method described above if the blood was made to flow immediately from the cut artery into the alcohol. On the other hand, if the shed blood was allowed to stand, the quantity of fibrin ferment increased up to the time of coagulation, and was present in quantity in the serum. Schmidt believed that the ferment was formed in shed blood from the disintegration of the leucocytes, and this belief was corroborated by subsequent histological work. It was shown in microscopic preparations of coagulated blood that the fibrin threads often radiated from broken-down leucocytes-an appearance that seemed to indicate that the leucocytes served as points of origin for the deposition of the fibrin. When the blood-plates were discovered a great deal of microscopic work was done tending to show that these bodies also are connected with coagulation in the same way as the leucocytes, and serve probably as sources of fibrin ferment. In microscopic preparations the fibrin threads were found to radiate from masses of partially disintegrated plates; and, moreover, it was discovered that conditions which retard or prevent coagulation of blood often serve to preserve the delicate plates from disintegration. At the present time it is generally believed that there is derived from the disintegration of the leucocytes and blood-plates something that is necessary to the coagulation of blood, but there is some difference of opinion as to the nature of this substance and whether it is identical with Schmidt's fibrin ferment. Pekelharing thinks that the substance set free from the corpuscles and plates

is a nucleo-proteid, but that this nucleo-proteid is not capable of acting upon fibrinogen until it has combined with the calcium salts of the blood. According to his view, therefore, fibrin ferment, in Schmidt's sense, is a compound of calcium and nucleo-proteid. Lilienfeld has shown by chemical reactions that blood-plates and nuclei of leucocytes contain nucleo-proteid material which in all probability is liberated in the blood-plasma by the disintegration of these elements after the blood is shed. Lilienfeld contends, however, that solutions of fibrin ferment prepared by Schmidt's method do not contain any nucleo-proteid material, and that, although the liberation of nucleo-proteid material is what starts normal coagulation of blood, nevertheless so-called fibrin ferment is something entirely different from nucleo-proteid. In this point, however, his results are contradicted by the experiments of Pekelharing and of Halliburton, who both find that solutions of fibrin ferment prepared by Schmidt's method give distinct evidence of containing nucleo-proteid material. We may conclude, therefore, that the essential element of Schmidt's fibrin ferment is a nucleo-proteid compound. The nature of the action of the ferment on fibrinogen is quite undetermined. As was mentioned before, only a portion, and apparently a variable portion, of this fibrinogen appears as fibrin after clotting is completed. Along with the fibrin a new proteid fibrin globulin makes its appearance in the serum. This fact has suggested the view that perhaps the fibrin ferment acts after the manner of the digestive ferments by causing hydrolytic cleavage of the fibrinogen, that is, causes the fibrinogen molecule to take up water and then dissociate into two parts, fibrin and fibrin globulin. Hammarsten, however, is inclined to believe that the reaction is of a different nature, resembling more the change that occurs in the heat coagulation of proteids. According to this suggestion, the ferment causes a molecular rearrangement of the fibrinogen, resulting in the formation of fibrin, most of which is deposited in an insoluble form, while a smaller part, after suffering a still further alteration, appears as fibrin globulin.

Intravascular Clotting.-Clotting may be induced within the bloodvessels by the introduction of foreign particles, either solid or gaseous-for example, air—or by injuring the inner coat of the blood-vessels, as in ligating. In the latter case the area injured by the ligature acts as a foreign surface and probably causes the disintegration of a number of corpuscles. The clot in this case is confined at first to the injured area, and is known as a "thrombus." Intravascular clotting more or less general in occurrence may be produced by injecting into the circulation such substances as leucocytes obtained by macerating lymph-glands, extracts of fibrin ferment, solutions of nucleo-albumins of different kinds, etc. According to the theory of coagulation adopted above, injections of these latter substances ought to cause coagulation very readily, since the blood already contains fibrinogen, and needs only the presence of ferment to set up coagulation. As a matter of fact, however, intravascular clotting is produced with some difficulty by these methods, showing that the body can protect itself within certain limits from an excess of

ferment in the circulating blood. Just how this is done is not positively known, but there is evidence that it may be due mainly to a defensive action of the liver. Delezenne' states that when blood-serum is circulated through a liver it loses its power of inducing coagulation in a coagulable liquid, that is, probably its contained fibrin ferment is altered or destroyed. It seems probable that this action of the liver may be of importance in the normal circulation in maintaining the non-coagulability of the blood in the living animal. Moreover, injection of leucocytes sometimes diminishes instead of increasing the coagulability of blood, making the so-called "negative phase" of the injection. To explain this latter fact, it may be said that leucocytes give rise on disintegration to a complex nucleo-proteid known as nucleo-histon. Nucleo-histon in turn is said to be broken up in the circulation, with the formation of a second nucleo-proteid, leuconuclein, that favors coagulation, and a proteid body, histon, that has a retarding influence on coagulation. The predominance of the latter substance may account for the "negative phase" under the conditions described.

Why Blood does not Clot within the Blood-vessels.-The reason that blood remains fluid while in the living blood-vessels, but clots quickly after being shed or after being brought into contact with a foreign substance in any way, has already been stated in describing the theories of coagulation, but will be restated here in more categorical form. Briefly, then, blood does not clot within the blood-vessels because fibrin-ferment is not present in sufficient quantities at any one time. Leucocytes and blood-plates probably disintegrate here and there within the circulation, but the small amount of ferment thus formed is insufficient to act upon the blood, and the ferment is quickly destroyed or changed, probably by an action of the liver as stated above. When blood is shed, however, the formed elements break down in mass, as it were, liberating a relatively large amount of nucleo-proteids, which, together with the calcium salts, produce fibrin from the fibrinogen.

Means of Hastening or of Retarding Coagulation.-Blood coagulates normally within a few minutes, but the process may be hastened by increasing the extent of foreign surface with which it comes in contact. Thus, moving the liquid when in quantity, or the application of a sponge or a handkerchief to a wound, will hasten the onset of clotting. This is easily understood when it is remembered that nucleo-proteids arise from the breaking down of leucocytes and blood-plates, and that these corpuscles go to pieces more rapidly when in contact with a dead surface. It has been proposed also to hasten clotting in case of hemorrhage by the use of ferment solutions. Hot sponges or cloths applied to a wound will hasten clotting, probably by accelerating the formation of ferment and the chemical changes of clotting. Coagulation may be retarded or be prevented altogether by a variety of means, of which the following are the most important:

1. By Cooling.-This method succeeds well only in blood that clots slowly-for example, the blood of the horse or the terrapin. Blood from 1 Travaux de Physiologie, Université de Montpellier, 1898.

hese animals received into narrow vessels surrounded by crushed ice may be kept fluid for an indefinite time. The blood-corpuscles soon sink, so that this method is an excellent one for obtaining pure biood-plasma. The cooling probably prevents clotting by keeping the corpuscies intact.

2. By the Action of Neutral Suits.-Blood received at once from the bloodvessels into a solution of such neutral salts as sodium sulphate or magnesium sulphate, and well mixed, will not clot. In this case also the corpuscles settle slowly, or they may be centrifugalized, and specimens of plasma can be obtained. For this purpose horse's or cat's blood is to be preferred. Such plasma is known as salted plasma" it is frequently used in experiments in coagulation—for example, in testing the efficacy of a given ferment solution. The best salt to use is M2SO, in solutions of 27 per cent.: 1 part by volume of this solution is usually mixed with 4 parts of blood; if cat's blood is used a smaller amount may be taken-1 part of the solution to 9 of blood. Salted plasma or salted blood again clots when diluted sufficiently with water or when ferment solutions are added to it. How the salts prevent coagulation is not definitely known-possibly by preventing the disintegration of corpuscles and the formation of ferment, possibly by altering the chemical properties of the proteids.

3. By the Action of Albumose Solutions-Certain of the products of proteid digestion, peptones and albumoses, when injected into the circulation retard clotting for a long time. For injection into dogs one uses 0.3 gram to each kilogram of animal. If the blood is withdrawn shortly after the injection, it will remain fluid for a long time. The peptone solutions, on the contrary, have no effect on the clotting of blood if added to it in a glass outside the body. This curious action of peptone has been much discussed. In an interesting paper by Delezenne, referred to on the previous page, two important facts are brought out that furnish the author a basis for a credible theory of the anticoagulating effect of the injections. It has been shown, in the first place, that the peptone injections cause a marked and rapid destruction of blood leucocytes. Secondly, that if blood and peptone are circulated together through a living liver the mixture not only does not clot itself, but will prevent clotting when added to freshly drawn blood. The hypothesis to explain these facts and also the action of peptone on coagulation is that the peptone by destroying the leucocytes sets free nucleo-proteid and histon (see p. 61); the former of these by forming fibrin ferment would promote coagulation, but in passing through the liver it is destroyed or neutralized in some way, and the histon left in the blood is the substance that retards the clotting. It would be desirable, in connection with this hypothesis, if chemical proof were furnished that histon is present in the blood after peptone injections.

4. Many other organic substances have an effect similar to peptone when injected into the circulation or in some cases when mixed with shed blood. For example, extracts of leech's head, extracts of the muscle of the crayfish, the serum of the eel, a number of bacterial toxins, and many of the soluble