THE CHEMICAL MECHANISMS OF DEFENCE 1085 



serum, proteins, ferments, albumoses, partaking of the same property. All 

 such substances are classed together as antigens. Thus human serum 

 injected into a rabbit produces in the rabbit's serum some body which will 

 give a precipitate when mixed with human serum even in minute traces. 

 This predpitin formation is specific, so that it may be used as a test for the 

 origin of any unknown specimen of serum. In the same way rennet fer- 

 ment when injected gives rise to the production of an anti-rennin which will 

 neutralise the action of this ferment on milk. Antigens are all colloidal in 

 character and probably optically active. Ordinary drugs do not give rise 

 to the formation of anti-bodies, a necessary condition being apparently 

 some similarity in the molecular structure of the antigen to the proto- 

 plasm of the animal on which it acts and to which it becomes linked by 

 its haptophore group. 



CYTOLYSINS. The bacteria of tetanus and diphtheria cannot exist in 

 the body, infection by them being limited to a surface or abscess cavity. 

 When a disease involves infection of the tissues themselves by living micro- 

 organisms, somewhat more complicated mechanisms are brought into play 

 for the defence of the organism. We have already seen that normal blood 

 serum may exert a paralytic or destructive action on bacteria. Light has 

 been thrown on the factors involved in this destruction by a study of the 

 phenomenon of hcemolysis, i.e. the destruction of red blood corpuscles. 

 Normal goat's serum may be mixed with the red blood corpuscles of the 

 sheep without -any injury to the latter. If however sheep's corpuscles, 

 previously washed in normal saline, be injected at intervals of a few days 

 into a goat, the goat's serum is found to have acquired the power of rapidly 

 dissolving the red blood corpuscles. This hsemolytic power can be 

 proved by mixing the serum and the washed blood corpuscles together 

 and allowing the mixture to stand in a narrow tube. The corpuscles 

 rapidly sink to the bottom, leaving the colourless serum above, unless 

 haemolysis has occurred, in which case the serum will be of a transparent 

 red colour. If the haemolytic serum be heated to 55 C. it is found to have 

 lost its power of dissolving sheep's corpuscles. This power is at once restored 

 if to the heated serum be added any normal blood serum, even of the sheep 

 itself. It seems therefore that two substances are involved in the haemolysis, 

 namely, (a) a substance present in most normal sera which is destroyed at 

 a temperature of 60 C. and has been called the complement, and (b) a 

 substance present in the serum only as a result of the previous injection of 

 some species of red blood corpuscle, which is resistant to the action of heat 

 and is called the amboceptor. The reason for these names will be at once 

 apparent from the following experiment. Haemolytic goat's serum is 

 mixed with sheep's red blood corpuscles and the whole mixture kept at C., 

 at which temperature haemolysis is indefinitely delayed. After some time 

 the corpuscles are separated by means of the centrifuge. On testing the 

 supernatant fluid it is found to have no action on sheep's corpuscles, though 

 it still possesses the power of activating another specimen of serum which 

 has been heated. The serum separated from the corpuscles has thus lost the 



