In considering the mode of occurrence of thrombosis, it is necessary to refer to the Coagulation of blood. According to the views which are identified with the name of Alexander Schmidt (although Andrew Buchanan anticipated many of his results) three agents are necessary for coagulation, the fibrinogen, the fibrino-plastic substance or para-globulin, and the fibrine ferment. The two former unite to form fibrine in the presence of the latter. The fibrinogen is dissolved in the blood-plasma; the paraglobulin, at least chiefly, and the ferment, entirely, reside in the white corpuscles. More recent observations seem to have modified this view, and it now appears that coagulation occurs by the conversion of the dissolved fibrinogen into insoluble fibrine, this being effected by means of the fibrine ferment which resides in the leucocytes. A third element however is still necessary, namely, the presence of a salt of calcium without which coagulation does not occur. It is only by destruction of the white corpuscles that the ferment is set free; so long as the white corpuscles circulate in the blood and remain alive fibrine cannot form. Fibrine will form when the conditions are such that the white corpuscles are no longer preserved alive. Fibrine, it will thus be seen, is the result of a chemical process, and the resulting albuminous substance, the fibrine, is not a vital structure but a dead chemical compound. In order to the preservation of the white corpuscles they nfust not be exposed to the contact of dead matter. You may keep blood fluid for a long time if you simply ensure that it is in contact with living tissue. Lister long ago showed that if an artery be ligatured in two places, and cut out while full of blood, it may be hung up and the blood will remain fluid for some days. Within the body if a vessel be ligatured carefully in two places, the middle portion remaining in connection with the living tissues, then the blood may be kept fluid from twelve to fifteen days. When blood is exposed to a perfectly smooth surface, even if it be of dead matter, it does not coagulate readily. Thus a piece of glass in the circulating blood does not induce coagulation, and blood in a vessel whose internal surface is smeared with oil does not readily coagulate.

It appears from the observations of Kausehenbach that the cells of lymphatic glands when treated with water yield, like leucocytes, the ferment necessary for coagulation. Most other cells probably possess a similar power. Foa and Pellacani assert that when fresh brain substance is treated with water and filtered, the filtrate injected into the jugular vein of rabbits induces rapid coagulation.

Bizzozero asserts that it is the blood-plates (see p. 64) and not the leucocytes which have to do with coagulation. He would adopt the view given above of this process, substituting the blood-plates for leucocytes. This view, however, has not been supported by subsequent observers. In the observations of Wooldridge, a deposition of small granules was observed apart from ordinary coagulation, and he regarded these as composed of a form of Fibrinogen (A Fibrinogen). Lowit, on the other hand, holds that the granules are composed of globulin, and are due partly to disintegration of the leucocytes and partly to precipitation from the plasma. It may be concluded that, besides the process of coagulation described above, and ' without such coagulation, the blood is capable of depositing solid granules which have in the mass the characters of fibrine, and give the same reaction as fibrine with Weigert's stain (Lowit). This occurs very rapidly whilst blood is cooling, and can scarcely fail to occur when blood from the living body is examined in the ordinary way by placing a drop on a glass slide and covering it with a thin cover glass. It occurs also by mechanical interference with the blood.

In connection with this whole subject the alleged rapid disappearance of leucocytes when the blood is shed is of interest. Lowit asserts that by examining the blood removed from the living body in warm oil on a warm stage he prevents, at least in part, the disintegration of the leucocytes and the appearance of the disklets. His enumerations by this method seem to show that the leucocytes are much more numerous in the normal blood than they appear in an ordinary preparation, the numbers in the latter case being 80 per cent, fewer than in the former. These observations confirm the views held by Cohnheim, Schmidt, and others.