MODE OF ACTION OF SALTS 129 



one, and bears no relation to the chemical nature of the ions. Thus the effect 

 of Ca** cannot be distinguished from Ba". These effects, especially in the case 

 of the multivalent ions, are manifest even in very dilute solutions. (6) They 

 may affect the nature of the solvent in which they are dissolved — the so-called 

 lyotropic effect. This action has been studied by several workers, prominent 

 among whom are Hofmeister (1888, 1889) and Freundlich (1903). Hofmeister, 

 working with a number of neutral salts, found that these could be arranged in a 

 definite order relative to their action on the coagulation of colloids, and on other 

 physical properties of proteins. Freundlich, studying the effect of electrolytes on 

 the compressibility, surface tension, solubility, viscosity, and other properties of 

 proteins, was likewise able to arrange them in a definite order. He concluded 

 that the main effect was exerted not on the proteins directly, but upon the sol- 

 vent ; modifications in the solvent thus affected the proteins. For this reason 

 he spoke of the salt effects as " lyotropic " effects ; his series of salts hence bears 

 the name " lyotropic series " in distinction to the " Hofmeister series." Both 

 series, however, are similar in many respects. Holm and Sherman (1921) and 

 numerous other workers have found a general concordance between the ionic 

 series of Hofmeister or of Freundlich and the stimulative or toxic action of the 

 salts of these series, (c) They may operate through a specific influence, which 

 is more intimately connected with the chemical properties of the ions. Thus 

 sodium and potassium, though having the same electrical charge, and exercising 

 very similar lyotropic effects, are yet totally different in their action on heart 

 muscle. As this action is shown by solutions so dilute that undissociated mole- 

 cules are nearly absent, we know that this difference must be attributed to some 

 specific or chemical property of the ions. 



Other explanations of ionic action have been put forward, some of them 

 modifications of the ones already given. Amongst these may be mentioned 

 Loeb's (1899, 1900) hypothesis of ion-protein combination, Mathews' (1904a, h, 

 1905, 1906) conception of ionic potential, and Zwaardemaker's (1918, 1919-20) 

 radioactivity hypothesis. For these and for further information on this subject 

 the reader is referred to an admirable summary by Falk (1923) — a summary 

 which has been freely drawn on in this section. 



There is one further point. We have treated the action of anions and of 

 cations separately ; whether both are sometimes required simultaneously for the 

 production of the bactericidal effect is not at all clear. It seems probable that 

 in a salt such as HgClg, in which a toxic cation is united to a weakly toxic anion, 

 almost the entire action of the salt must be referred to the Hg-ion ; in a salt such 

 as KjMnaOg, on the other hand, in which a weakly toxic cation is united to a power- 

 ful anion, the action must be referred to the permanganate ion. 

 Summary of Salt Action. 



(1) There is a certain concentration for nearly all salts which stimulates bac- 

 terial growth ; this concentration is generally very low. 



(2) There is, for nearly all salts, a limit beyond which the stimulating action 

 passes over into a toxic action ; on the whole, the higher the concentration, the 

 more evident does the toxic action become. 



(3) The toxic effect of univalent salts can be neutralized by the addition in 

 suitable proportions of a divalent salt. In most instances, too, it is possible for 

 a univalent salt to neutralize the toxic action of a bivalent salt. This action is 

 known as the antagonistic action of salts. 



P.B. F 



