118 MNETEENTH REPORT. 



was then filtered off and the remaining filtrate eontaining the amino 

 aeids was freed from its alcohol by eva|)oration and subjected to the 

 Van Slyke method for the determination of amino acids. The results 

 showed an increase in amino nitrogen over a control which was treated 

 with the same volume of .85% salt solution. 



Since the majority of theories of anaphylaxis involve proteolysis, it 

 has occurred to us, too, to investigate the anaphylotoxin production from 

 the same standpoint. Is proteolysis the cause, is it concomitant with or 

 is it related in any way to the production of anaphylotoxin.'* We have 

 made use of the Van Slyke method for the determination of amino 

 acids of proteolysis and have used agar as the anaphylotoxin inducing 

 agent. Our technique otherwise was the same as that of Bordet and Zunz 

 which I have explained. 



RAT SERUM ANAPHYLOTOXIN. 



By tlie use of rat serum treated with agar we have been able to 

 produce anaphylotoxin in five minutes. Bordet and Zunz by using guinea 

 pig serum obtained their toxin after two hours' incubation. By pro- 

 duction of the maximum toxicity in the shortest period of time we sought 

 to eliminate the chances for the hydrolysis of the serum proteins by the 

 action of enzymes or any other cause. We were thus able to show 

 repeatedly that rapid toxification of rat serum with agar was accom- 

 panied by a decrease in amino acids over controls of untreated serum 

 incubated for the same length of time, rather than an increase as one 

 might expect. 



The use of guinea pig serum gave us the same results. We are able 

 to ))roduce a potent toxin from normal guinea })ig serum with 1.5 

 minutes' incubation. In every case in many series of exj)eriments a 

 similar decrease in amino acids was found to accompany toxification. 



Thinking that if the time of incubation were increased we miglit find 

 an increase in the amino acids of hydrolysis we incubated both rat and 

 guinea pig serum with agar for varying lengths of time from three to 

 21 hours and observed that after the preliminary drop there was a 

 gradual rise in amino acids but at no time even after 2i hours' incuba- 

 tion did the quantity of amino acids reach that of tlie control untreated 

 serum incubated for the same length of time. The control, too. showed tlie 

 same gradual increase in amino aeids ])ointing to the possibility of serum 

 autolysis M'itli incubation. Normal guinea pig serum treated with agar 

 which is removed at once without incubation shows the same fairly marked 

 decrease. We are at a loss to explain this j)henomenon and for the 

 present attribute it to adsorption by agar surface. An attempt has been 

 made to recover tin- amino acids thus adsorlied but without success. 



