and this in turn is formed from the plasma of the blood which is transuded, or, according to some physiologists, secreted, through the vessel-walls.

Paraglobulin, which belongs to the group of globulins, exhibits the general reactions characteristic of the group. As stated above, it is completely precipitated from its solutions by saturation with MgSO. It is incompletely precipitated by saturation with common salt (NaCl). In neutral or feebly acid solutions it coagulates upon heating to 75° C. Hammarsten gives its percentage composition as-C 52.71, H 7.01, N 15.85, S1.11, O 23.24. Schmiedeberg gives it a molecular composition corresponding to the formula CH182N3SO 38 +H2O. According to Faust,' the precipitate of paraglobulin usually obtained with MgSO, contains a certain amount of an albuminoid body, glutolin, which he believes to be a constant constituent of blood-plasma. Paraglobulin occurs in blood, in lymph, and in the normal and pathological exudations. The amount of paraglobulin present in blood varies in different animals. Among the mammalia the amount ranges from 1.78 per cent. in rabbits to 4.56 per cent. in the horse. In human blood it is given at 3.10 per cent., being less in amount, therefore, than the serum-albumin. It will be seen, upon examining the tables of composition of the blood-plasma and blood-serum of the horse. (p. 51), that more of this proteid is found in the serum than in the plasma. This result, which is usually considered as being true, is explained by supposing that during coagulation some of the leucocytes disintegrate and part of their substance passes into solution as a globulin identical with or closely resembling paraglobulin. The figures given above show that a considerable amount of paraglobulin is normally present in blood. It is reasonable to suppose that, like serum-albumin, this proteid is valuable as a source of nitrogenous food to the tissues. It is uncertain, however, whether it is used by the tissues directly as paraglobulin or is first converted into some other form of proteid. It is entirely unknown, also, whether its value as a proteid supply is in any way different from that of serum-albumin. The origin of paraglobulin remains undetermined. It may arise from the digested proteids absorbed from the alimentary canal, but there is no evidence to support such a view. Another suggestion is that it comes from the disintegration of the leucocytes (and other formed elements) of the blood. These bodies are known to contain a small quantity of a globulin resembling paraglobulin, and it is possible that this globulin may be liberated after the dissolution of the leucocytes in the plasma, and thus go to make up the normal supply of paraglobulin. The fact remains, however, that at present the origin and the special use of the paraglobulin are entirely unknown.

Fibrinogen is a proteid belonging to the globulin class and exhibiting all the general reactions of this group. It is distinguished from paraglobulin by a number of special reactions; for example, its temperature of heat coagulation is much lower (56° to 60° C.), and it is completely thrown down from its solutions by saturation with NaCl as well as with MgSO,. Its most important and distinctive reaction is, however, that under proper conditions it gives 1 Faust, Inaugural Dissertation, Leipzig, 1898.

rise to an insoluble proteid, fibrin, whose formation is the essential phenomenon in the coagulation of blood. Fibrinogen has a percentage composition, according to Hammarsten, of-C 52.93, H 6.90, N 16.66, S 1.25, O 22.26; while its molecular composition, according to Schmiedeberg, is indicated by the formula C108H162N30SO34

Fibrinogen is found in blood-plasma, lymph, and in some cases, though not always, in the normal and pathological exudations. It is absent from bloodserum, being used up during the process of clotting. It occurs in very small quantities in blood, compared with the other proteids. There is no good method of determining quantitatively the amount of fibrinogen, but estimates of the amount of fibrin, which cannot differ very much from the fibrinogen, show that in human blood it varies from 0.22 to 0.4 per cent. In horse's blood it may be more abundant-0.65 per cent. As to the origin and the special physiological value of this proteid we are, if possible, more in the dark than in the case of paraglobulin, with the exception that fibrinogen is known to be the source of the fibrin of the blood. But clotting is an occasional phenomenon only. What nutritive function, if any, is possessed by fibrinogen. under normal conditions is unknown. No satisfactory account has been given. of its origin. It has been suggested by different investigators that it may come from the nuclei of disintegrating leucocytes (and blood-plates) or from the dissolution of the extruded nuclei of newly-made red corpuscles, but here again we have only speculations, that cannot be accepted until some experimental proof is advanced to support them.

Coagulation of Blood.-One of the most striking properties of blood is its power of clotting or coagulating shortly after it escapes from the bloodvessels. The general changes in the blood during this process are easily followed. At first shed blood is perfectly fluid, but in a few minutes it becomes viscous and then sets into a soft jelly which quickly becomes firmer, so that the vessel containing it can be inverted without spilling the blood. The clot continues to grow more compact and gradually shrinks in volume, pressing out a smaller or larger quantity of a clear, faintly yellow liquid to which the name blood-serum has been given. The essential part of the clot is the fibrin. Fibrin is an insoluble proteid that is absent from normal blood. In shed blood, and under certain conditions in blood while still in the blood-vessels, this fibrin is formed from the soluble fibrinogen. The deposition of the fibrin is peculiar. It is precipitated, if the word may be used, in the form of an exceedingly fine network of delicate threads that permeate the whole mass of the blood and give the clot its jelly-like character. The shrinking of the threads causes the subsequent contraction of the clot. If the blood has not been shaken during the act of clotting, almost all the red corpuscles are caught in the fine fibrin meshwork, and as the clot shrinks these corpuscles are held more firmly, only the clear liquid of the blood being squeezed out, so that it is possible to get specimens of serum containing few or no red corpuscles. The leucocytes, on the contrary, although they are also caught at first in the forming meshwork of fibrin, may readily pass out into the serum in the later stages of clot

ting, on account of their power of making amoeboid movements. If the blood has been agitated during the process of clotting, the delicate network will be broken in places and the serum will be more or less bloody-that is, it will contain numerous red corpuscles. If during the time of clotting the blood is vigorously whipped with a bundle of fine rods, all the fibrin will be deposited as a stringy mass upon the whip, and the remaining liquid part will consist of serum plus the blood-corpuscles. Blood that has been whipped in this way is known as "defibrinated blood." It resembles normal blood in appearance, but is different in its composition: it cannot clot again. The way in which the fibrin is normally deposited may be demonstrated most beautifully under the microscope by placing a good-sized drop of blood on a slide, covering it with a cover-slip, and allowing it to stand for several minutes until coagulation is completed. If the drop is now examined, it is possible by careful focussing to discover in the spaces between the masses of corpuscles many examples of the delicate fibrin network. The physiological value of clotting is that it stops hemorrhages by closing the openings of the wounded blood

vessels.

Time of Clotting.-The time necessary for the clot to form varies slightly in different individuals, or in the blood of the same individual varies with the conditions. It may be said in general that under normal conditions the blood passes into the jelly stage in from three to ten minutes. The separation of clot and serum takes place gradually, but is usually completed in from ten to forty-eight hours. The time of clotting shows marked variations in different animals; the process is especially slow in the horse and the terrapin, so that coagulation of shed blood is more easily prevented in these animals. In the human being also the time of clotting may be much prolonged under certain conditions in fevers, for example. This fact was noticed in the days when bloodletting was a common practice. The slow clotting of the blood permitted the red corpuscles to sink somewhat, so that the upper part of the clot in such cases was of a lighter color, forming what was called the "buffy coat." The time of clotting may be shortened or be prolonged, or the clotting may be prevented altogether, in various ways, and much use has been made of this fact. in studying the composition and the coagulation of blood as well as in controlling hemorrhages. It will be advantageous to postpone an account of these methods for hastening or retarding coagulation until the theories of coagulation have been considered.

Theories of Coagulation.-The clotting of blood is such a prominent phenomenon that it has attracted attention at all times, and as a result numerous theories to account for it have been advanced. Most of these theories possess simply an historical interest, and need not be discussed in a work of this character, but some reference to older views is unavoidable for a proper presentation of the subject. To prevent misunderstanding it may be stated explicitly in the beginning that there is at present no perfectly satisfactory theory. Indeed, the subject is a difficult one, as it is intimately connected with the chemistry of the proteids of the blood, and it may be said that a complete understanding

of clotting waits upon a better knowledge of the nature of these proteids. It is possible that at any moment new facts may be discovered that will alter present ideas of the nature of the process. In considering the different theories that have been proposed there are two general facts that should always be kept in mind: first, that the main phenomenon that a theory of coagulation has to explain is the formation of fibrin; second, that all theories unite in the common belief that the fibrin is derived, in part, at least, from the fibrinogen of the plasma.

Schmidt's Older Theory of Coagulation.-The first theory that gained general acceptance in recent times was that of Alexander Schmidt. It was proposed in 1861, and it has served as the basis for all subsequent theories. Schmidt held that the fibrin of the clot is formed by a reaction between paraglobulin (he called it "fibrinoplastin ") and fibrinogen, and that this reaction is brought about by a third body, to which he gave the name of fibrin ferment. Fibrin ferment was believed to be absent from normal blood, but to be formed after the blood was shed. Further reference will presently be made to the nature of this substance. Schmidt was not able to determine its naturewhether it was a proteid or not-but he discovered a method of preparing it from blood-serum, and demonstrated that it cannot be obtained from blood immediately after it leaves the blood-vessels, and that consequently it does not exist in circulating blood, in any appreciable quantity at least. Finally, Schmidt believed that a certain quantity of soluble salts is necessary as a fourth "fibrin factor."

Hammarsten's Theory of Coagulation.-Hammarsten, who repeated Schmidt's experiments, demonstrated that paraglobulin is unnecessary for the formation of fibrin. He showed that if a solution of pure fibrinogen is prepared, and if there is added to it a solution of fibrin ferment entirely free from paraglobulin, a typical clot is formed. This experiment has since been confirmed by others, so that at present it is generally accepted that paraglobulin takes no direct part in the formation of fibrin. Hammarsten's theory was that there are two fibrin factors, fibrin ferment and fibrinogen, and that fibrin results from a reaction between these two bodies. The nature of this reaction could not be determined, but Hammarsten showed that the entire fibrinogen molecule is not changed to fibrin. In place of the fibrinogen there is present after clotting, on the one hand, fibrin representing most of the weight of fibrinogen (60-90 per cent.), and, on the other hand, a newlyformed globulin-like proteid retained in solution in the serum, to which proteid the name fibrin-globulin has been given. Hammarsten supposed that although paraglobulin took no direct part in the process, it acted as a favoring condition, a greater quantity of fibrin being formed when it was present. Later experiments' indicated that this supposition was incorrect, and that paraglobulin may be eliminated entirely from the theory. The theory of Hammarsten, which is perhaps generally accepted at the present time, is incomplete, however, in that it leaves undetermined the nature of the ferment Frederikse: Zeitschrift für physiologische Chemie, Bd. 19, 1814, S. 143.

and of the reaction between it and fibrinogen. The aim of the newer theories has been to supply this deficiency.

1

Schmidt's Theory of Coagulation.-In a volume containing the results of a lifetime of work devoted to the study of blood-coagulation, Schmidt has modified his well-known theory. His present ideas of the direct and indirect connection of the proteids of the plasma with the formation of fibrin are too complex to be stated clearly in brief compass. He classifies the conditions necessary for coagulation for coagulation as follows: (1) Certain soluble proteids— namely, the two globulins of the blood-as the material from which fibrin is made. Schmidt does not believe, however, that paraglobulin and fibrinogen react to make fibrin, but believes that fibrinogen is formed from paraglobulin, and that fibrinogen in turn is changed to fibrin. (2) A specific ferment, fibrin ferment, to effect the changes in the proteids just stated. He proposes for fibrin ferment the distinctive name of thrombin. (3) A certain quantity of neutral salts is necessary for the precipitation of the fibrin in an insoluble form. The Relation of Calcium Salts to Coagulation.-It has been shown by a number of observers that calcium salts take an important part in the process of clotting. This fact was first clearly demonstrated by Arthus and Pages, who found that if oxalate of potash or soda is added to freshly-drawn blood in quantities sufficient to precipitate the calcium salts, clotting will be prevented. If, however, a soluble calcium salt is again added, clotting occurs promptly. This fact has been demonstrated not only for the blood, but also for pure solutions of fibrinogen, and we are justified in saying that without the presence of calcium salts fibrin cannot be formed from fibrinogen. This is one of the most significant facts recently brought out in connection with coagulation. We know that fibrinogen when acted upon by fibrin ferment produces fibrin, but we now know also that calcium salts must be present. What is the relation of these salts to the so-called "ferment"? The most explicit theory proposed in answer to this question we owe to Pekelharing. Pekelharing's Theory of Coagulation.-Pekelharing succeeded in separating from blood-plasma a proteid body that has the properties of a nucleoalbumin. He finds that if this substance is brought into solution together with fibrinogen and calcium salts, a typical clot will form, while nucleoalbumin alone, or calcium salts alone, added to fibrinogen solutions, cause no clotting. His theory of coagulation is that what has been called "fibrin ferment" is a compound of nucleo-albumin and calcium, and that when this compound is brought into contact with fibrinogen a reaction occurs, the calcium passing over to the fibrinogen and forming an insoluble calcium compound, fibrin. According to this theory, fibrin is a calcium compound with fibrinogen or with a part of the fibrinogen molecule. This idea is strengthened by the unusually large percentage of calcium found in fibrin ash. The theory supposes also that the fibrin ferment is not present in bloodplasma—that is, in sufficient quantity to set up coagulation—but that it is formed 1 Zur Blutlehre, Leipzig, 1893.

2

2 Untersuchungen über das Fibrinferment, Amsterdam, 1892.