54 THE HUMAN BODY. 



lined by serous membranes, contain fibrinogen, fibrinoplas- 

 tin, and salts, but little or no ferment, therefore they do 

 not coagulate spontaneously or only imperfectly and slowly. 

 But if a little blood serum be added to one of these liquids, 

 coagulation quickly occurs. 



Artificial Clot. If serum be slightly diluted with water 

 and kept ice-cold while a stream of carbon dioxide gas is 

 passed through it for some hours, a white precipitate is 

 thrown down which contains fibrinoplastin and the fibrin 

 ferment. This precipitate after washing may be dissolved 

 in cold water containing the merest trace of caustic potash. 

 If the liquid moistening a serous cavity be treated in a 

 similar way a precipitate is formed, containing fibrinogen 

 instead of the fibrinoplastin, and but little of the ferment. 

 If this precipitate be washed and dissolved and the solution 

 be added to the solution of the blood-serum precipitate, 

 no clot is formed; but if about one per cent of sodic car- 

 bonate or other neutral salt be added to the mixture, then 

 it clots. This shows the necessity of the salts, which is 

 perhaps better proved in another way. If serum be put in 

 a dialyzer (see Physics) with distilled water on the other 

 side of the membrane, all the salts will gradually pass out 

 from the serum into the water: as the last portions of 

 them pass out, the fibrinoplastin and ferment, which are 

 "colloids" (bodies which do not readily dialyze), are pre- 

 cipitated; they may be redissolved by the addition of a- 

 trace of caustic potash. Similarly the salts may be re- 

 moved from the liquid obtained from a serous cavity, and 

 the precipitated fibrinogen redissolved. If these solutions 

 be now mixed no clot is formed; but if the salts which have 

 been dialyzed out, or an equivalent portion of other neu- 

 tral salts, be added to the mixture, it will clot. 



The Fibrin Ferment. The activity of the ferment is 

 proved as follows: If serum be diluted with a fih^e bulk of 

 water and then carbon dioxide gas be passed through it, 

 fibrinoplastin will be precipitated, with little or none of 

 the ferment. If this fibrinoplastin be added to the fibrin- 

 ogenous liquid from a serous cavity it will not cause it to 

 clot, or only very slowly, according as no fibrin ferment or 



