98 The Nature of Precipitin Reactions 



stance, but that they adhere in different proportions to its constituents. 

 In support of this assumption they state that trypsin-treated albumin 

 gives a reaction with precipitins when all the albumin has been broken 

 up as indicated by the biuret reaction. They conclude that the 

 precipitin-formation, due to immunization with albumins of egg-white, 

 is independent of albumins, and that it depends upon a substance which 

 is difficult to separate from albumins in the process of chemical cleaning, 

 consequently it cannot be a specific process due to the action of albumins 

 of egg-white. The immunifying substance, which they wrongly style "pre- 

 cipitogen\" as well as the precipitin they conclude are not albuminous, 

 and consequently the biological test is of no use in the determination of 

 albumins. They found anti-egg serum to resist the action of hot 0"5 "/o 

 nitric acid. These observers and others have found the precipitins in 

 different antisera to be precipitated by alcohol. 



The rapidity of the reaction between precipitin and precipitable 

 substance certainly points to the existence of great affinity between the 

 interacting substances. 



In relation to the p7'ecipitahle substance, Halban and Landsteiner 

 (25, III. 1902, p. 475) have found that serum-albumin, rendered salt-free 

 by dialysis and subsequently heated to 100° C, reacted but slightly to 

 precipitins, though when boiled 15 minutes it contained scarcely less 

 albumin (as understood by chemists and physiologists) than before. 

 It would therefore appear that precipitation constitutes a reaction with 

 definite chemical groups (within the albuminous molecule ?) like the 

 agglutination reaction, etc. Linossier and Lemoine (18, IV. 1902) and 

 Leclainche and Vallee (25, i. 1901) find that the action of antiserum is 

 not comparable to that of other reagents such as heat or nitric acid. 

 This is proved by the fact that the precipitin does not always give 

 reactions proportional to the amount of albumin present. Antisera are 

 more active towards globulins than nitric acid, less so towards serum- 

 albumin. With urine which contains much of the latter substance an 

 antiserum may give no reaction. 



Eisenberg (v. 1902, p. 306) notes that heated albumin is more 

 alkaline than native, and that alkalized albumin is not precipitable. 

 He found (p. 307) albumin to be modified by contact with concentrated 

 solution of urea, added in the proportion of 4 : 1 to albumin solution, and 



1 A misnomer, as pointed out by Michaelis and Oppenheimer (1902, p. 341) for the 

 reason that it suggests a relation between precipitin and (" prccipitogen ") precipitable 

 substance such as exists, for instance, between pepsin and pepsinogen, in other words that 

 the " precipitogen " is a forerunner of precipitin, which it is not. 



