440 ILEMOLYSINS AND ALLIED BODIES 



its tissues. This cytolytic power may be increased in any animal 

 by the repeated injection into it, in non-lethal but gradually 

 increasing dosage, of these foreign cells. This increased cytolytic 

 power depends, not on an increase of the destructive agency, but 

 on the development in the tissue fluids of some substance which 

 increases the sensitiveness of the invading cell towards the 

 normally existing destructive agency of the tissues. 



Bordet's explanation of the facts of bacteriolysis and haemolysis 

 does not, however, satisfactorily or completely explain all the 

 observed phenomena of the formation of anti-bodies, (e.g. why 

 the sensitising substance should be so remarkably specific in its 

 nature, why there should be so great an increase of the substance 

 in immune blood, &c.). It is to Ehrlich and Morgenroth that 

 we owe a conception which seems to explain all the observed facts. 

 These workers have evolved a theory which has become well 

 known as the side-chain theory. This theory permits of consider- 

 able speculation, and by it a large number of possible conditions 

 can be predicted and all the observed facts of immunity explained. 

 It is of fundamental importance for us at this stage, therefore, to 

 thoroughly master the details of this theory, and the experiments 

 on which it is founded. 



Let us first of all consider the simplest conception of the side- 

 chain theory. This applies to antitoxin formation. If a bouillon 

 culture of the diphtheria or tetanus bacillus be filtered through 

 a Pasteur-Chamberland filter, the filtrate will contain tetanus- or 

 diphtheria-toxin, and will, if injected into susceptible animals, 

 cause death. If, however, the toxin be injected at intervals of a 

 few days apart, at first in sub-lethal doses, and the doses gradually 

 increased, the animal will become immune. A dose of the toxin, 

 many times in excess of what would originally have killed the 

 animal, comes to have no action on it. Moreover, the serum of 

 such an immune animal will, if injected into another normal 

 animal, protect the latter also against subsequent injection of the 

 toxin. There must, therefore, be something developed in the 

 blood of the immunised animal which can neutralise the toxin. 

 This is called antitoxin. Ehrlich says that toxin and antitoxin 

 unite with one another in definite quantities, and in multiples 

 of these quantities ; x that cold retards the union ; and that con- 

 centrated solutions react quicker than dilute ones. 



1 The work of Danysz, Craw, &c., show this is not so, and afford evidence in 

 favour of Bordet's theory. (Editor's Note.) 



