of Cerium, Lanthanum, and Didymium. 113 



And the composition of the metal itself is — 



II. 



Didymium 99*181 



Silicon ...... 0-368 



Iron 0-339 



Aluminium 0*112 



100-000 

 From this analysis it appears that the 0*9071 grm. of metal 

 which was employed contained 0*8997 grm. of didymium, and 

 gave, after removal of the impurity contained in the 1*0626 

 grm. of oxide, 1*0488 grm. of the pure oxide of didymium. 

 The composition of the oxide of didymium, as found by direct 

 analysis, is therefore — 



III. 

 Didymium . . . 85*78 per cent. 

 Oxygen .... 14*22 „ 



100*00 

 The percentage quantity of oxygen in didymium oxide de- 

 duced from P. T. Cleve's * analysis of the didymium sulphate, 

 according to its saturation-capacity, is 14*04, agreeing very 

 nearly with that directly found. The agreement would be 

 still closer had it been possible to determine the amount of the 

 lanthanum in the didymium. 



Let d = the amount of didymium contained in the unit 

 weight of the metal under investigation, 

 k = that of the silicon, 

 e = that of the iron, 

 a = that of the aluminium ; 



further, let the specific heat of these bodies be respectively s d , 

 sic, $e, and s a , and let S = the observed specific heat of the im- 

 pure metal ; then 



ds d + ks k + es e + as a = $, 

 and hence 



S—ks k —es e —as a _ s . /g\ 



d 

 by substituting the following values, 



^=0-99181, $ = 0-04653, 

 £ = 0-00368, ** = 0-175, 

 ^=0-00339, s e =0*114, 

 a = 0'00112, s a = 0'214, 



in this equation, we obtain for the specific heat of pure didy- 

 mium the value 0*04563. 



* Abliandl. d. Schwedischen Akad. vol. ii. No. 6. 



Phil. Mag. S. 5. Yol. 3. No. 16. Feb. 1877. I 



