REy\CTING POTENCY 26] 



tion of the suspensions. In these expciinienis it was lound that 

 the reacting potency of a <>i\en precipitate suspension could be 

 (}uantitati\ely measured. As in tiie experiments witli tlie phenom- 

 enon of local skin reactivity to bacterial filtrates, there were also 

 obser\ed nattnal fluctuations in susceptibility of rabbits employed. 

 Since the niunber of resistant animals rarely exceeded 25 per cent, 

 the tise of groups of 3 to 4 rabbits allowed fairly accurate titra- 

 tions. 



The titrations demonstrated that precipitates deri\ed at various 

 times from mixtines of the same ingredients and in the same 

 proportions, may differ in reacting potency. It became also evi- 

 dent that the amount of precipitate present in a given suspension 

 has no direct bearing on its reacting potency. Thus, a prepara- 

 tion of only slight tmbidity (/.r., dilution 1:50) was potent, ^vhilst 

 several highly ttnbid preparations (/.t'., dilutions 1:5) were totally 

 inactive. Similar observations ^vere obtained with titrations car- 

 ried out in a some\vhat difTerent manner. Suspensions of precipi- 

 tates were diluted in 0.85 per cent NaCl solution to a ttubidity 

 standard (MacFarland nephelometer) of 500 million bacteria per 

 1 c.c. The dose was 2 c.c. per kilo of body weight. The mixtines 

 consisted of one part of anti-luunan horse sertnn and one part of 

 human serimi diluted 1:4. Two different batches of the anti- 

 hiunan horse serum Avere employed. With one batch of the horse 

 serimi a suspensi(jn of standard tmbidity and dilution 1:5 elicited 

 severe reactions and dihuions 1:10 ga\e negati\e results. \V'ith 

 another batch of anti-luunan horse sertnn a suspension of the pre- 

 cipitate of standard ttubidity and suspension t\vo and four times 

 more concentrated gave no reactions. 



REACTING POTENCY OF PRECIPITATES AND SUPERNATANT FLUmS 



DERIVED FROM COMPLETELY AND PARTIALLY 



PRECIPITATED MIXTURES 



The reacting potency of a serum precipitate is apparently not 

 due to the mechanical effect of an inert colloidal suspension in 

 the blood stream but to some toxic qtiality associated with the 

 precipitate. This, indeed, is supported by a ninnber of variotis 

 observations discussed tmder this and subseqtient headings of 

 this chapter. 



Mixttnes of sertnn with antisertnn in proportions giving a 

 maximinn ]:)recipitation yielded active precipitates and inactive 

 supernatant fluids. ^V'here precipitations ^vere incomplete {i.e., in 



