SALTS OF THE GLOBULINS 



123 



potassium hydroxide, so that ammonium globulinate does not 

 suffer an appreciable hydrolytic dissociation. This stronger 

 combination with alkalis as compared with acids reflects the 

 very acidic character of globulin. 



The quantity of alkali hydroxide required to dissolve i gm. 

 of globulin is of the order of o-i millimol, and this gives the high 

 equivalent weight of 10,000 for these salts of the globulin. 

 Double this quantity of alkali is necessary to neutralise the 

 globulin when phenol phthalein is used as indicator, thus giving 

 an equivalent weight of 5,000. The globulinate prepared in 

 this way, when its conductivity is determined, behaves as if the 

 globulin ions were divalent, thus making the molecular weight 

 10,000. Hardy, using Sackur's somewhat dubious method of 

 calculation, obtains the value 5 for the valency, giving the 

 altogether too high value of 25,000. 



The conductivity of strong acids and bases, which is high 

 owing to the great mobility of the H> and OH' ions, falls off 

 when neutralisation takes place. Thus the ratio of the con- 

 ductivity of the salt and of the equivalent concentration of free 

 acid or alkali is less than unity. On the other hand, it can 

 become equal to i, or even greater, if the salts of weak acids 

 or bases are compared with the original acid or alkali. 



Ammonium globulinate conducts the current much better 

 than an equivalent solution of ammonium hydroxide ; the ratio 

 of the conductivities is 2-3. On the other hand the corre- 

 sponding quotient for sodium globulinate and sodium hydroxide 

 is 0-9. 



In the results obtained by Hardy for the conductivity of 

 globulin when acid or alkali is added, the measurements were 

 made at such great dilutions that they are not unexceptionable 

 either with regard to the large water-error or to the disturbing 

 effects of hydrolysis. They lead, therefore, to values which are 

 contradictory, and in particular far too high for the conductivity 

 of the globulinate ion. One can, however, deduce the mobility 

 of the globulinate ion, v G \ ob from some of Hardy's experiments 

 at a moderate dilution in the following way. The molecular 

 conductivity /x at a dilution of V litres at 18 is given by the 

 table on p. 124. 



