1889.] On Auto-infection in Cardiac Disease. 311 



used for injection. I am therefore inclined to explain the fact that 

 the lymph does not normally poison the blood because it runs into the 

 blood slowly whilst the blood circulates rapidly. In a normal state, 

 therefore, the conditions which must exist for a fibrinogen intoxica- 

 , ion do not prevail. 



I have above used the term '* poison the blood " ; it will be advan- 

 tageous for me to explain this expression. 



The admixture of fibrinogen and blood may obviously affect the 

 latter, by causing it to clot or by preventing its clotting (vide previous 

 papers),* but it produces other changes than these which are not so 

 directly perceptible. The nature of these changes will be seen from 

 the following : If in a normal dog the femoral vein be ligatured 

 there is no obvious effect produced, i.e., there is no oedema of the leg. 

 If, however, some solution of fibrinogen be injected into the circula- 

 tion through the jugular vein and the femoral be then ligatured, the 

 effect produced is most pronounced, and is as follows : either the 

 most extensive and rapidly developing simple cedema of the leg occurs 

 or an enormous haemorrhage "per diapedesin" takes place throughout 

 the tissues of the limb ; or the two are combined there is haemorrhage 

 and oedema. 



The injection of fibrinogen,f then, in addition to the obvious effects 

 of clotting or delay in clotting, produces a totally disturbed relation- 

 ship between the blood and the vascular wall, since, after the injection, 

 a slight mechanical disturbance to the circulation causes a greatly 

 increased exudation of the fluid of the blood, or this associated with a 

 free passage of the red corpuscles. The tendency the injection has 

 to cause haemorrhage I have already pointed out in a previous publi- 

 cation, the fact that it produces a simple but severe and sudden 

 cedema is new. Now, to produce this altered state of the blood, 

 leading to cedema, the same conditions of admixture of blood and 

 fibrinogen are necessary, i.e., the admixture must be rapid. I will 

 illustrate this by an experiment. 



Experiment 3. 



Used the NaCl fluid of thymus free from cells. 



Dog I. Weight 17 Ibs. 12 c.c. of solution rapidly injected into 

 the jugular. Right femoral vein tied close to Poupart ligament. 

 Dog killed the next day. The portal system thrombosed. The whole 



* " Intravascular Clotting," ' Roy. Soc. Proc.,' 1886; " Beitrage zur Frage der 

 Gerinnung," * du Bois-Reymond, Archiv,' 1888; "Ueber Schutzimpfung auf 

 Clu-mischein Wege," ' du Bois-Reymond, Archiv,' 1888. 



t The fibrinogen used to produce this effect may be lymph fibrinogen, tissue 

 fibrinogen, or certain varieties of blood fibrinogen. 



I Wooldridge, " On Hsemorrhagic Infarction of the Liver," ' Pathol. Soc. Proc.,' 

 1888. 



