304 Drs. D. A. Welsh and H. G. Chapman. On the [Apr. 30, 



2-5 mm. Hence the precipitum resulting from the precipitin reaction, 

 notwithstanding its looser consistence, cannot be explained as a derivative 

 of the homologous proteid alone. 



After 48 hours 04 c.c. of the clear superfluid from each tube was trans- 

 ferred to each of two clean tubes, A and B. To each A tube was added 

 0-00012 gramme egg white in - 5 c.c. saline ; to each B tube - 01 gramme 

 antiserum in 0'5 c.c. saline. The precipitates in 48 hours are given in 

 Table VII. 



Table VII. — Secondary Interactions. Experiment IV. 



No. 



Preoipita in A series 

 ( + egg white). 



Precipita in B series 

 ( + antiserum) . 





mm. 



mm. 



1 



0-5 



1-5 



2 



1 -5 



0-5 



Therefore, both precipitin and homologous proteid coexisted uncombined 

 in the clear superfluids ; the precipitin was largely neutralised in the first 

 tube, but not much in the second ; the homologous proteid was not notice- 

 ably exhausted in any tube. The last result is the most important for our 

 present purpose. 



Tertiary Interactions. — Having noted that the secondary interactions 

 invariably revealed the presence of homologous proteid not noticeably 

 exhausted, whether the proteid were of the nature of egg white or of the 

 nature of serum, we submitted some of the superfluids of the secondary 

 interactions to the influence of a fresh supply of proteid and of antiserum. 

 The addition of proteid revealed the presence or absence of precipitin 

 according as it had not or had previously been neutralised. The addition 

 of antiserum invariably revealed the presence of unexhausted homologous 

 proteid by the deposition of precipitates which were not noticeably smaller 

 than if the same quantities had interacted originally. It is unnecessary 

 to detail these observations, since similar results are more strikingly 

 displayed in the following experiments : — 



Exhaustion Experiments. — These observations were so unexpected and so 

 much at variance with accepted views regarding the precipitation of the 

 homologous proteid that we instituted a series of experiments to try if, by 

 successive additions of antiserum to the clear superfluids remaining after 

 each interaction, we could exhaust the homologous proteid. As the following 

 experiments show, we are not in a position to say whether the proteid is or 



