prk.sidp:nt"s address — section l. 277 



Another experiment leads to a similar conclusion. It can be 

 ascertained what is the minimum amount of complement necessary 

 to cause complete haemolysis of a certain small quantity of sensi- 

 tized red corpuscles. Now, take a much larger quantity of comple- 

 ment, but, before mixing it with the corpuscles, add to it an 

 amount of anti-complement not nearly sufficient to ne'utralize it 

 compleCely. 



On Ehilicb's hypothesis the mixture should contain a large 

 amount of free fully-active complement, able to ha^molyze com- 

 pletely several times the small unit dose of corpuscles. As a matter 

 Of fact, it will be found that this will be quite unable to hsemolyze 

 the same quantity of corpuscles in unit time,, for, hcinr/ iniiformi// 

 attenuated, it works slowly. On the other hand, it will liberate 

 a much larger quantity of haemoglobin from a large dose of cor- 

 puscles than the minimum dose of fully active complement could 

 do. In other words, as Bordet puts it, the one is superior by the 

 factor of activity, the other by the factor of quantity. 



An observation of Buchner's with tetanus toxin and antitoxin 

 is of interest. Weight for weight, a guinea pig is much more 

 susceptible to poisoning with tetanus-toxin than a mouse. Thus, 

 a guinea pig weighing twenty times as much as a mouse will be 

 poisoned by ten times the lethal dose for a mouse. Now, if to a 

 fairly large quantity of tetanus-toxin some antitoxin is added in 

 just a sufficient quantity to render the mixture tolerable to a 

 mouse, it will be found that the same dose of the mixture injected 

 into a guinea pig will kill it. This is inexplicable if one imagines 

 the antitoxin completely neutralizing a proportion of tbe toxin, 

 but leaving a quantity free and active. It is easily understood if 

 one imagines the whole of the toxin attenuated uniformly, for so 

 attenuated it is ineffective against a mouse but effective against a 

 guinea pig. 



Another observation which has to be noted is that there is a 

 power of selection, of greater or less affinity, in adsorption, as in 

 purely chemical combinations, and that one element may be able 

 thus to displace another. Thus, Gengcu (1908) has shown that a 

 precipitate of barium sulphate, whilst having a strong adsorptive 

 affinity for mucin, has an even stronger one for the albuminoids 

 of serum. The addition of mucin to the barium sulphate precipi- 

 tate causes an aggregation or agglutination of the fine particles. 

 Now, on adding a little serum the mucin is displaced, and the 

 agglutinated particles again disperse. In a similar fashion, as 

 Morgenroth has shown, sensitized red corpuscles retain the sensi- 

 tizing substance tenaciously if suspended in saline solution ; but if 

 a further quantity of red corpuscles from the same species, this 

 time iin sensitized, is added, the sensitized ones give up to them 

 part of the sensitizing substance. A simple experiment will illus- 

 trate the point admirably. Place in a watch, glass of water a disc 

 of filter paper previously stained fairly deeply with methylene blue 



