434 Mr. C. C. Hutchinson on the Separation and 



obtained, and to reprecipitate, so as to thoroughly eliminate 

 every trace of zinc. 



The method of separation described by Fresenius, and 

 usually given by other authorities, is also open to the ob- 

 jection of not yielding concordant results, even though every 

 precaution be observed in the process and the quantity of 

 the reagents be carefully adjusted. This method is based 

 upon the insolubility of the hydrated oxide of cadmium in an 

 alkaline solution containing tartaric acid. To the hydrochloric 

 acid solution of the two, as near as possible neutral, tartaric 

 acid is added ; sufficient sodium or potassium hydrate is then 

 added until the solution is distinctly of an alkaline reaction. 

 If this solution be boiled for about two hours, the hydrated 

 oxide of cadmium separates out on cooling, the zinc being left 

 in solution. After filtration, the hydrated oxide is ignited 

 and weighed as the oxide (CdO). The zinc may be precipitated 

 as the sulphide by ammonium sulphide, or titrated with sodium 

 sulphide. 



Two examples may be given, showing the mutual varia- 

 bility in estimations which should each have yielded the same 

 result. In each case the determination of the amount of cad- 

 mium contained in a mixed precipitate of zinc and cadmium 

 was the object in view. 





grin. 



erm. 



N T o. 1 gave CdO . 



. . 0-0805 or Cd . 



. . 0-0704 



„ 2 „ CdO . . 



, . 0-0615 „ Cd . 



. . 0-0538 



Differences 0-0190 0-0166 



The next example given shows the result of experiments 

 upon precipitates of the same nature, but containing a larger 

 amount of each substance. 



grm. grin. 



No. 3 gave CdO . . . 0-194 or Cd . . . 0-1697 

 „ 4 „ CdO . . . 0-1165 „ Cd . . . Q-1019 



Differences 0'0775 0*0678 



The differences here shown, taken in relation with the com- 

 paratively small amount of the metal present, are serious, and 

 indicate the method as unreliable. The subsequent zinc-esti- 

 mation in all of these cases showed that in no instance was the 

 whole of the cadmium present precipitated ; the error, there- 

 fore, relatively to the total cadmium present, was even larger 

 than above shown. This was amply confirmed by other ex- 

 periments, in all of which the solubility of the hydrated oxide 

 of cadmium, and its subsequent precipitation with the zinc 

 as sulphide, was inevitably found ; in many instances the 



