326 LOCAL TLSSIIE REACIIVLrV 



tency. lIuiiKiii and guinea j)ig whole blood and scia j)icseiil a 

 notable exception yielding irregularly potent provocative factors 

 when used in large anioinits. Inasmtich as both man and guinea 

 pig are known to contain the Forssnian antigen and also be- 

 cause almost invariably a strong agglutination of rabbit blood 

 cells may be obtained with human serum, the reacting potency of 

 these sera may possibly be related to this type of in vivo antigen -|- 

 antibody interaction. It seems, however, that the Forssman antigen 

 alone in gtiinea pig and himian blood cannot be held entirely 

 responsible for the reacting potency inasmuch as dog and horse 

 blood belonging to the same group as himian and guinea pig sera 

 produce no reactions in prepared skin sites. Thus, it appears that 

 the reacting potency of whole blood and sera of man and guinea 

 pig may be related to still another interaction not explained as yet 

 between the normal antibodies with the tissue and cell antigens 

 of the rabbit. This field requires fmther investigation. 



In 1914, Ledingham produced a condition resembling general- 

 ized purptna and presenting other clinical features of hemor- 

 rhagic diathesis by injection of guinea pig anti-platelet serum into 

 guinea pigs. By varying the route of administration of anti-platelet 

 serum and adjusting the dose injected, Roskam (1931) obtained 

 local hemorrhagic lesions. In an extensive series of studies, Bedson 

 (1922) tested the effect of variotis antisera prepared by immu- 

 nization ^vith various blood elements and tissues (namely, plate- 

 lets, red blood cells, leucocytes, blood serum, spleen, endothelial 

 leucocytes from the peritoneum, bone marrow, and lymph 

 glands) . Of these sera, anti-platelet and anti-spleen sera alone 

 were capable of producing purpura. The purpura was the result 

 of the action of the platelet antibody alone and apparently took 

 place independently of any hemaggliuination. The experimental 

 evidence presented by Bedson and Roskam strongly suggests that 

 two main factors are responsible for the production of hemor- 

 rhage: 1. toxic action on the endothelium of the vessels, and 2. 

 removal of platelets from the circidation. 



The effect of the pinpura-producing sera is apparently totally 

 mirelated to the phenomenon under discussion in this mono- 

 graph. Thus, serum of guinea pigs immunized with the rabbit 

 spleens produced local purpuric and necrotic lesions on intra- 

 dermal and subcutaneous injection into rabbits. Skin sites thus 

 prepared did not react to the intravenous injection of toxic bac- 

 terial filtrates. No enhancement of the local reaction Avas pro- 



