90 ATOMIC WEIGHT DETERMINATIONS. 



that Rose dealt with a compound containing a large amount 

 of the corresponding tantalium chloride. {Foggend. Ann., 

 104., 1858, 439.) 



Rose ; Rammelsberg : 94 (O = 16). 



Rose analysed the oxychloride, but did not recognize it 

 as an oxychloride. Rammelsberg calculated the atomic 

 weight from Rose's figures and found that the highest chlo- 

 rine contents corresponds to an atomic weight of 94. 

 Rose's salt must have been nearh^ pure as there is no cor- 

 responding tantalium compound. {Poggend. Ann., 136, 

 1869, 353.) 



R. Hermann : 104.8 (0 = 16). 



Hermann deduces this value from analyses of a number 

 of chlorides and sodium salts. The formulas which he 

 gives these compounds are complicated, unlikely, and un- 

 supported b}' evidence. Marignac has shown that Her- 

 mann's salts contained tantalium. [Erdmann^s Journ. fiir 

 Prak. Chem., 68, 1856, 73.) 



C. W. Blomstrand: 95(0 = 16). 



Blomstrand made three determinations of the chlorine 

 contents of the pentachloride, getting 64.712 per cent., ex- 

 treme difference, 0.32. He also made eleven determina- 

 tions of the niobium in the same compound, weighing it as 

 niobic acid. 100 chloride gave in mean 49.794 acid. The 

 atomic weight calculated from the chlorine contents is 

 96.67; from the niobic acid, 96.16. Blomstrand also made 

 experiments on sodium niobate which led him to the con- 

 clusion that the most probable number is 95. {Gmelin- 

 Kraut, 2, part 2, 73 ; Acta Univ. Lund., 1864.) 



C. Marignac : 94 (O = 16). 



Determined from a number of analyses of potassium 

 fluoniobate containing two atoms of potassium. The com- 

 pound was decomposed by sulphuric acid with which it 

 was evaporated to dryness. The residue was leached with 

 water, the filtrate evaporated and the potassic sulphate 

 melted and weighed. The sulphuric acid remaining with 

 the niobic acid was driven off by heat and the acid weighed. 

 The salt being readily soluble and crystallizing well, can 

 easily be purified from all substances except titanium which 

 Marignac knows no way of separating or determining. 



