1883.] 



On the Atomic Weight of Titanium. 



45 



The total amount of the silver chloride thus obtained was 

 69*25983 grins. : it was formed from 53T3881 grms. of silver added in 

 the preceding series. The weight of the silver chloride thus obtained 

 from a known weight of silver and of titanium tetrachloride, not only 

 affords an additional value for Ti, but it also serves as a rigorous 

 check on the accuracy of the work, for if the ratio of Ag to AgCl 

 is found to be the same as that obtained by the direct union of 

 chlorine and silver, it at once disposes of the possibility of error due 

 to the co-precipitation of titanic acid, and is a guarantee of the purity 

 of the silver employed — 



Now 52 -13881 : 69 -25983= Ag : AgCl = l : 1 '3284. 



Stas found from seven experiments in which, in the aggregate, 

 969*3548 grms. of silver were found to give 1287'7420 grms. of silver 

 chloride, that — 



Ag : AgCl=l : 1 -32845. 

 The two ratios, it will be seen, are almost identical. 



Series III.— TiCl 4 : Ti0 2 . 



The tetrachloride was decomposed by water, and the solution 

 evaporated to dryness and strongly heated. 



Chloride taken 



XIV 6 '23398 



XV 8 -96938 



XVI 10-19853 



XVII... 6-56894 



XVIII 8 '99981 



XIX 8 -32885 



49 -29948 



Titanic oxide. 



2 -62825 



3 -78335 



4 -30128 

 2-77011 

 3 -79575 

 3 -51158 



20 -79032 



Ti (H = l). 



47 

 48 

 47 

 47 

 47 

 47 



93 

 00 

 95 

 96 

 98 

 94 



47 -970 



All the experiments which were made, either for the estimation of 

 the chlorine or the titanic oxide, are given with the exception of two ;. 

 one of these was made upon a small quantity of material, and was 

 considered as merely preliminary ; the other miscarried and was not 

 persevered with. 



In order to enable an opinion to be formed as to the agreement 

 among the several observations, I have calculated the value of Ti 

 afforded by each experiment. But probably the most accurate values 

 for the atomic weight would be deduced from the aggregate weights 

 of the tetrachloride, silver, silver chloride, and titanic oxide, respec- 

 tively. Inasmuch as it may be presumed that the employment of large 

 quantities of material would tend to increase the accuracy of the 

 result, by calculating the final values from the aggregate weights 



