

1894.] Oxygen upon the Coagulability of the Blood in Vivo. 287 



free carbonic acid. The power which these solutions have of in- 

 libiting blood coagulation is inferred to be associated with this 

 sroperty; (g) thrombosis of the pulmonary vessels occurs in doers 

 rhen they are caused to breathe atmospheric air in which the whole 

 litrogen has been replaced by C0 2 ; Qi) after burns, the venous blood 

 found to contain a great excess of carbonic acid. The coagulability 

 such blood is abnormally high. 



It will be observed that the experiments which have been the 

 ibject-matter of the present communication entirely confirm the 

 inclusions which had already been arrived at by entirely different 

 lethods by Mathieu and Urbain. I have not in any systematic 

 lanner controlled the observations upon which the conclusions of 

 lese observers were based. I have, however, had many incidental 

 spportunities of confirming their observations with respect to the 

 Iterations of blood coagulability which are conditioned by raising 

 >r cooling the general body temperature. On the other hand, I 

 ive not, in the very few cases I have examined for it, observed 

 the occurrence of pulmonary thrombosis as an effect of a simple 

 ise of carbonic acid tension in the blood, but I have, in one 

 briking experiment, seen a rabbit whose blood coagulability had 

 )een increased by the administration of calcium chloride die in- 

 intaneously from universal intravascular coagulation, when it was 

 ipplied with an atmosphere which was surcharged with carbonic acid. 



Appendix. Selected Protocols. 



The appended protocols are to be read from left to right, and then 

 3k in a zig-zag manner, following the dotted lines from right to 

 ft. In accordance with the fact that the method of coagulability 

 determinations which was employed is an approximal and not an 

 absolute one, two data are given for each coagulability determination, 

 lese data (longest interval during which the blood was observed to 

 smain in a tube unclotted, and shortest interval which the blood was 

 Hind to require for complete occlusion of a tube) are entered in sepa- 

 ite columns. Where only a single entry appears on the protocols this 

 indicative of a lacuna in the observations. Thus, when an entry 

 ippears in the first column only (as, for instance, in the case of the 

 jcond coagulability determination on the last protocol on the list), it 

 to be understood that the last of the coagulation tubes which were 

 appropriated to the particular coagulability determination was found 

 liquid when tested after the interval noted in the protocol. Simi- 

 irly, when an entry appears in the second column only, as, for 

 istance, in the second determination on the first protocol on the list, 

 is to be understood that all the tubes were found completely 

 clotted, although the testing of the tubes was not deferred beyond 

 ie fifty seconds noted on the protocol. 

 VOL. LT. X 



