86 JOURNAL OF THE 



various preparations, and I reg^ard them as'showing- con- 

 clusively that a neutral chloride of zirconium can be 

 prepared and dried. 



Analyses of this chloride ^ave the following- percent- 

 ag-es of ZrO^: 



52.7(» 52.78 52.63 



Experiments have already been beg^un with a view of 

 utilizing* this b(xly in a series of experiments looking" to 

 a revision of the atomic vveig'ht of zirconium. 



In connection with this subject it may be well to 

 mention some improvements in the method of purifying" 

 zirconium chloride. (See Journal of Analytical and 

 Applied Chemistry, 5, 551). 



In the first place the separation from silica by evapora- 

 tion to dryness is not complete. It is impossible to heat 

 this chloride to the necessary temperature without such 

 a decomposition as will render the zirconium chloride 

 also insoluble. It is best then to make this preparation 

 as thoroug-h as possible by heatingf, then to chang-e the 

 chloride into oxide by ig-nition, and to treat this several 

 times with hydrofluoric acid until the trace of silica is 

 all driven off. This silica is too small in amount to 

 interfere with ordinary uses but would have to be re- 

 moved where perfect purity was demanded. 



Ag-ain, where the hydroxide is dissolved in dilute hy- 

 drochloric acid, or contained so much water that the 

 acid was g-reatly diluted b}^ it, it will be found that 

 more or less of a white insoluble powder will form on 

 evaporation as recomfnended on a water-bath and on 

 subsequent treatment with boiling- strong" hydrochloric 

 acid. By a careful arrangfement of g-lass wool in a 

 hot water funnel the dissolved chloride can be filtered 



