1911.] Serum Reactions in Pulmonary Tuberculosis. 379 



seem that some qualitative change, displaying itself by increased anti- 

 complementary effect, took place, which is not associated with those 

 quantities in antigen which are concerned with specific antigen-serum 

 attraction or negation. 



The following protocols (Table II) seem to support this conception, although 

 they are not conclusive. Indeed, it is difficult to realise all the conditions 

 necessary to prove this, nor is it perhaps essential. The sediment (protocol, 

 Table II) was prepared by precipitating the stock solution at 0° C. and 



Table II.— 19.12.10. 





Haemolysis. 







A.E. IV. (1—10). 



1 



c 











5 











25 





Almost complete + . 



1 







' S. 312—0 -3 c.c. 











Complete. 







A.E. IV. Sediment (1—10). 



1 



c 





None. 







7 











5 





Almost none + . 







25 







1 







,' S. 312—0 -3 c.c. 









5 





» +■ 



1 







\ S. 1354— o"'2 c.c. 



Complete. 







5 









A.E. IV. Filtrate (] 



-10). 



1 



c 





None. 







7 











5 





Almost none. 







25 





» » + • 



1 







' S. 312—0 -3 c.c. 













Complete. 



1 







\ S. 1354— o"-2 c.c. 







5 





Almost complete. 



re-dissolving this material in an amount of alcohol-ether equal to the amount 

 of stock employed. The filtrate is the stock solution minus the precipita- 

 tion in the cold. The controls to complement showed that barely two 

 units of fixative were available. Unfortunately it was at this time 

 impossible to make observations with sensitiser sera. Sera 312 and 1354 

 had the day before given inhibitive reactions with another antigen. 



(6) Inhibitive Reaction and its Relation to Time. — The procedure has been 

 to incubate the first three quantities for one hour and to give the customary 



VOL. LXXXIV. — B. 2 F 



