190 



Dr. G. Dreyer and Mr. J. S. C. Douglas. [Nov. 16, 



we see that with a constant amount of bacteria and an increase in the total 

 amount of agglutinin present, C not only reaches a limit value at a certain 

 concentration, but after this point has been reached actually decreases 

 whilst the total amount of agglutinin present further increases, until at 

 a given coiicentration no diminution in the agglutinin content of the serum 

 can be traced as a result of its treatment with the bacterial emulsion ; 

 indeed, there seems in some cases even to be a tendency to a slight increase 

 in the agglutinating strength of the serum, though not, in this series of 

 experiments, of such degree as to be beyond the range of experimental 

 error. 



It is difficult to say how this phenomenon is to be explained, but, that it 

 cannot be a question of the presence of so-called " agglutinoids " (Eisenberg 

 and Volk) in the serum, the existence of which had been rendered most 

 unlikely by the previous experiments of Dreyer and Jex-Blake, is clearly 

 seen from the following fact. 



If a serum is first treated with great quantities of bacteria it is found by 

 afterwards using it for absorption experiments that not only is the attain- 

 ment of a limit value not prevented, but that even the decrease in the 

 actual size of C with increasing concentration of T is still evident 

 (Experiment 9, Table X). This, however, could not occur if, as is said to be 

 the case, the agglutinoids have a greater avidity for bacteria than the 

 agglutinins have. A previous treatment of the serum by bacteria would 

 free it from all, or the greater part of, such agglutinoids, so that the 

 limiting value for C found after they have presumably been greatly 

 diminished or got rid of cannot in any way be attributed to them. 



The correctness of the statement that this phenomenon is not due to 

 " agglutinoids " will be further proved by the absorption experiments under- 

 taken with fresh agglutinating serum in both a heated and an unheated 

 condition, referred to later in this paper. 



In our opinion the phenomenon of a decrease in tlie value of C with 

 increasing concentration of the serum is most likely caused by some obscure 

 alterations in the surface tension due to change in tlie concentration of the 

 all)unien or of the different salts, or in the viscosity of the Huid, etc. 



Such a limit value in C, and even an actual decrease in the size of C, 

 may be reached, not oidy in the absorption processes of agglutinin by their 

 own lioniologouH l)acteria, but may also occur if tlie agglutinin is acted on by 

 non-specific bacteria, as, for example, a Coli agghUiiiiii l)y an emulsion of 

 typhoid bacilli (see Experiment 2, Table III). 



We liave also ol)(aiiic(l similnr results by treating an agglutinin in various 

 dilutions with constant amount oi' animal chiircoal, as will hv. described 



