SIGNIFICANCE OF PHENOMENON 379 



thronil)osis. Evidence iiulicating that the lesions of the phenome- 

 non are associated with manifestations of altered vascular response 

 is demonstrated by the follo^ving experiments: 



An area of skin prepared by the injection of bacterial filtrate 

 responds in a normal manner to various stimidi, such as mechani- 

 cal irritation, ajjplication of heat (50° to 70° c.) , massage, 

 adrenalin, histamine, calciiun chloride, cooling by ether, cocaine, 

 etc. (Kielano^vski and Selzer, 1934/;) . 



After the pro\ocative injection of the active principles is made 

 intra\'enously, there is manifest a modified response of the blood 

 vessels of the tissues at sites distant from the local reaction. This 

 is best demonstrated by the appiication of suction, ptdling of 

 hair, shaving with a dtdl razor, etc. The normal skin sites thtis 

 treated develop petechial hemorrhages which, microscopically, 

 consist of capillary dilatation and extravasation of erythrocytes. 

 The enhanced fragility of the capillaries in sites distant from the 

 local reaction may be demonstrated for a period of four to forty- 

 eight horns follo^ving the provocative injection of bacterial fil- 

 trates and depends some^vhat on the severity of the local lesion 

 (Sh^vartzman) . Gratia and Linz dre^v attention to the observa- 

 tion that in rabbits in ^vhich the phenomenon is elicited in one 

 ear there may appear hemorrhagic lesions in the non-prepared ear 

 following the intravenotis injection of the active principles. This 

 reminds one of the symmetrical distribution of the hemorrhagic 

 lesion in the Schoenlein-Henoch syndrome. 



The tendency to^vards the experimental hemorrhagic diathesis 

 cannot be correlated as yet, with blood changes. In the experi- 

 ments of Mintz, mentioned before, no distinct changes were ob- 

 served in coagulation and bleeding time. It appears, therefore, 

 that the changes observed are to be ascribed primarily to certain 

 modifications in the behavior of the blood capillaries themselves. 



It is also possible to demonstrate that other changes take place 

 in sites not directly affected by the local reaction which predis- 

 pose towards an exaggerated tendency to the formation of 

 thrombi. W^hen the phenomenon is elicited in the rabbit's ear by 

 combined intradermal and intravenous injection of the active 

 principles, sites distant from the local reaction frequently become 

 cyanotic. In stich sites it is possible to detect, microscopically, 

 thrombosis of the ventdes irot accompanied by inflammatory peri- 

 vasctdar reactions (p. 20) . 



Thus, the essential morphological features of the phenomenon 



