26 PROPERTIES OF ILEMOLYTIC SERA 



appreciable amount cannot be satisfactorily demonstrated. 

 In one experiment twenty doses of complement were added 

 to 1 c.c. of a 20 per cent, suspension of red corpuscles along 

 with sufficient immune-body, and after suitable time for com- 

 bination 1 c.c. of sensitized corpuscles was added. There was 

 no lysis of the latter to be noticed, although in the conditions 

 of experiment the separation of one-fifth of a dose of comple- 

 ment would have been detected. Accordingly, in this example 

 we can say that there was no dissociation of even 1 per cent 

 of the complement which had combined. Another oppor- 

 tunity of judging with regard to this question is given when 

 there has been a small quantity of complement added in 

 excess, so that a fraction of the sensitized corpuscles which 

 are added undergoes lysis. We can place the tubes aside 

 for twenty-four hours at room temperature, and observe 

 whether there is any marked increase of lysis, such as might 

 occur through the dissociation of the combined complement. 

 In the case of the immune-serum, rabbit v. ox, along with 

 guinea-pig's complement, I have not found any distinct 

 evidence of dissociation of complement by this method. 

 In other cases, for example in the case of the immune-serum, 

 rabbit v. guinea-pig along with rabbit's complement, we do 

 find that the lysis increases under the conditions mentioned. 

 This might be due to dissociation of complement, but it 

 might also be due to the presence of very weakly acting 

 complements which are present in the serum, the effects 

 of which are only seen after some time. 



It is sufficient to state that in most cases of haemolysis 

 the combination of complement is a firm one, though the 

 possibility of dissociation of small quantities in certain 

 cases is not to be denied. A similar statement applies to 

 the union of complement and bacteria by means of the 

 corresponding immune-body. In the case of the absorption 

 of a complement by a serum and its anti-serum (p. 143) 

 we sometimes meet with phenomena which suggest dis- 



