i8o THE BLOOD. 



as supposed by Lowit l and by Wright, 2 but by causing their accumula- 

 tion within the tissues (? capillary blood vessels). For a very short 

 time after their almost complete disappearance from the blood they 

 begin gradually to reappear, and in one experiment C. J. Martin found 

 that the full number had reappeared within as short a time as fifteen 

 minutes. 3 Moreover, if such disintegration, really took place one would 

 expect the coagulability of the blood to be visibly increased, from the 

 setting free of their nucleo-proteids, whereas it is actually diminished 

 or abolished. Nevertheless, those substances, such as snake venom, 

 nucleo-proteids, and colloids, which in larger doses produce intravas- 

 cular coagulation, may in part act by causing disintegration, if not of 

 leucocytes at least of red corpuscles which also contain nucleo-proteids. 

 That this occurs to some extent is shown by the fact that the serum is 

 usually tinged by haemoglobin. And even without actual disintegration 

 the permeability of the corpuscles may become altered, and nucleo- 

 proteid shed out. 



But there is another tissue upon which the reagent in question may 

 act, namely, the epithelial cells of the blood vessels. These are in all 

 probability composed of living protoplasm, and the reagents may either 

 cause them to shed out nucleo-proteid and so produce fibrin ferment, 

 or, by deleteriously affecting them, may cause them to react upon the 

 leucocytes which are passing along in contact with their inner surface, 

 and effect a discharge of nucleo-proteid from these cells. That snake 

 venom affects the blood vessels deleteriously is shown by the capillary 

 haemorrhages which are so frequently seen after poisoning by it, and 

 by the rapid effect it produces on the blood circulating in the mesentery, 

 if a little be applied to the surface of that membrane. 4 The same does 

 not, however, obtain with artificial colloids, nor with nucleo-proteids ; 

 although, with partial blocking of the portal vein, after injection of a 

 small dose, capillary haemorrhages have been found to occur in the liver. 5 



The evidence which we have had before us points to the following 

 conclusions regarding coagulation : 



1. That the coagulation of blood, i.e. the transformation of fibrinogen 

 into fibrin, requires for its consummation the interaction of a nucleo- 

 proteid (prothrombin) and soluble lime salts, and the consequent produc- 

 tion of a ferment (thrombin). 



2. That either nucleo-proteid is not present in appreciable amount 

 in the plasma of circulating blood, or that the interaction in question is 

 prevented from occurring within the blood vessels by some means at 

 present not understood. 



3. That the nucleo-proteid (prothrombin) appears and the interaction 

 occurs, as soon as the blood is drawn and is allowed to come into contact 

 with a foreign surface, the source of the nucleo-proteid being in all 

 probability mainly the leucocytes (and blood-platelets ?). 



4. That, under certain circumstances and conditions, either the 

 nucleo-proteid does not appear in the plasma of drawn blood, or it 

 appears, but the interaction between it and the lime salts is prevented 

 or delayed. 



1 "Stud. z. Physiol. u. Path. d. Blutea," Jena, 1892. 



2 Proc. Roy. Soc. London, 1893. vol. Hi. 



3 Loc. cit. 



4 Weir Mitchell and Reichert, " Researches upon the Venoms of Poisonous Serpents," 

 Smithson. Contrib. Knowl., Washington, vol. xxvi. 



5 Wooldridge, Trans. Path. Soc. London, 1888, p. 421. 



