99 



.052 



Calculating the ratio between CO 2 and GIG, we have for 

 the molecular weight of the latter, GIG = 25.220, ± .180. 

 The agreement between this result and the one previously 

 deduced from the sulphate is certainly very striking. 



Last of all and best of all we come to the determinations 

 recently published by Nilson and Pettersson.* These chem- 

 ists sought to use the sublimed chloride of glucinum, but 

 found it to contain traces of lime derived from a glass tube. 

 They finally resorted to the sulphate as the most available 

 salt for their purposes. This, which they write Gl2(SG4)3 

 I2H2G, and which we formulate as GISG4.4H2G, yields pure 

 glucina upon strong ignition. The subjoined percentages 

 of glucina were thus obtained : 



14.171 

 14.169 

 14.160 

 14.176 



Mean, 14.169, zh .0023 



Hence GIG = 25.048, and Gl = 9.085, ± .0055. If G = 

 16, Gl = 9.106. If SGg == 80, then Gl = 9.096. 



If the oxide is Gl^Og, then the value Gl = 9.085, ± .0055 

 becomes Gl = 13.628, ± .0082. 



It would be easy enough to combine this value for Gl 

 with those derived from the experiments of the investi- 

 gators previously cited, but it is hardly worth while. All 

 the other estimations have such high probable errors that 

 they would practically vanish from the general mean. 

 Their influence would hardly extend to the third decimal 

 place, and they may therefore be neglected. 



* Compt. Rend., gi, i68. 1880. 



