370 CARNEGIE INSTITUTION OF WASHINGTON. 



the time the complement-fixation reaction occurred the precipitin and 

 passive anaphylaxis reactions appeared, but were at first limited to the 

 homologous protein. 



Further experiments with various proteins strongly suggest that the 

 severity of the anaphylaxis reaction is proportional to the solubility 

 of the proteins in the peritoneal solution. Proteins like edestin, which 

 are precipitated on injecting into the peritoneal fluid, are so slowly re- 

 dissolved that the animal rarely reacts fatally, whereas very small quan- 

 tities of those vegetable proteins which are freely soluble in the peri- 

 toneal fluid cause fatal reactions. Thus, from 5 to 10 milligrams of 

 most of the vegetable globulins were required to produce severe intoxi- 

 cation, but from 0.05 to 0.10 milligram of the so-called ''vegetable pro- 

 teoses" caused severe reactions and 0.5 to 2.0 milligrams gave fatal 

 results. The alcohol-soluble proteins, gliadin and hordein, are more 

 toxic than any of the vegetable globulins of the series tested. Since 

 the proteoses formed by enzymatic or acid hydrolysis of native pro- 

 teins do not cause anaphylactic intoxication, it is probable that most of 

 the so-called "vegetable proteoses" should not be thus designated. 

 The high toxicity of these substances corresponds with that of the native 

 proteins and indicates that the "vegetable proteoses" have a similar, 

 highlj' complex chemical structure. 



We have also cooperated with Dr. Benjamin White and Dr. 0. T. 

 Avery in studying the immunity reactions of edestin and gliadin. 

 This study has shown that edestin, in even small amount, agglutinates 

 the red blood corpuscles of sheep or man, whereas gliadin does not. 

 The serum of rabbits immunized with edestin contains a precipitating 

 antibody for edestin, but not for gliadin. Edestin, in the presence 

 of edestin-immune serum, completely binds complement, but gliadin, 

 under these conditions, does not. In agreement with Dr. Wells's results 

 the minimum sensitizing dose of edestin was found to be 0.0000001 

 gm., injected intraperitoneally into guinea-pigs. When 0.0500 gm. of 

 edestin is subsequently injected intraperitoneally into animals thus 

 sensitized they die quickly. If the sensitizing dose was 0.0001 to 

 0.0050 gm., 0.5 milligram of edestin, given intravenously, produced 

 typical anaphylactic death in from two to six minutes. Guinea-pigs 

 sensitized with edestin gave no reaction when the globulins of the 

 squash-seed, castor-bean, or hazel-nut were intravenously injected, 

 but reacted with large doses of the globulin of the flax-seed. Guinea- 

 pigs born of a mother sensitized with edestin were also sensitized, but 

 to a less degree than the mother. Passive sensitization to edestin 

 appeared to be somewhat greater than that induced by active sensiti- 

 zation. Edestin hydrolyzed by the Vaughan method yielded a sub- 

 stance which caused an apparently true anaphylactic shock. Edestin 

 yields a toxic product which appeared to correspond to the anaphyla- 

 toxin of Friedberger. 



