272 THE POWER OF RESISTANCE TO EXTREMES 



acid character. Such reactions as these acquire their chief importance from 

 the insight of the protoplasmic constitution which they may render possible. 



The facts observed in the case of organic compounds suffice to show 

 that the physiological action is determined by the grouping of the component 

 atoms and atom-complexes, and is not the direct result of the properties of 

 the individual atoms. When a compound is split up, the physiological 

 action is the resultant of the sum of the properties of the simpler materials l . 

 The same applies when a substance undergoes dissociation, so that the elec- 

 trolytic salts of a poisonous metal may all exercise the same physiological 

 and chemical action when the acid ions are indifferent 2 . When the dissocia- 

 tion is incomplete, the undissociated portion may exercise a physiological 

 action of its own apart from that due to the dissociated ions. 



When a dissociating compound is innocuous, its ions must also be 

 non-poisonous. Experiments with the chlorides, sulphates, and nitrates of 

 potassium, sodium, and calcium have shown that both the metallic ions 

 (kations) and the acid ions (anions) are either quite innocuous or only very 

 feebly poisonous. Hence the poisonous character of potassium cyanide is 

 due to the anion cyanogen, and that of the chlorides, nitrates, and sulphates 

 of mercury and copper to the metallic ions, for all these salts are entirely 

 or nearly entirely dissociated when in considerable but still poisonous dilution. 

 The same principles apply to free acids in which hydrogen plays the part of 

 a metal, and to free alkalies in which hydroxyl replaces the acid ester (anion). 

 Since the anions and kations of the neutral salts of the above-mentioned 

 alkalies and alkaline earths are non-poisonous, it follows that the injurious 

 action of free acids is due to the hydrogen ions and that of alkalies to the 

 anion hydroxyl. 



In acids with innocuous anions the poisonous action is dependent upon 

 the degree of dissociation. Thus hydrochloric, nitric, and sulphuric acids 

 which undergo complete dissociation when largely diluted are extremely 

 poisonous, whereas the slightly dissociating acetic and propionic acids are 

 only feeble poisons. The combining avidity of such acids is also slight as 

 compared with those which readily dissociate. When the anion of an acid is 

 highly poisonous, as in the case of the cyanogen radicle of prussic acid, only a 

 small portion of its poisonous action is due to its properties as a weak acid. 



Similar relationships are shown by alkalies such as sodium and potas- 

 sium hydrate, which dissociate very largely in dilute solutions. Clark found 

 that the hydroxyl ion was more poisonous to fungi than that of hydrogen, 

 whereas the reverse was found to be the case by Kronig and Paul in bac- 

 teria, and by Kahlenberg and True in Phanerogams. It remains to be seen 

 whether this is due to differences in the specific resistance of the groups 



1 Special results may be produced by such conjoint action. 



3 Cf. Ostwald, Wiss. Grundlageu der analytischen Chemie, 1897, p. 44. 



