DiDYMIUM. 3 



Frekichs and Smith. Mr. Brauner then tried heating the basic nitrate 

 of didymiiim to incipient redness in a current of oxygen. On determi- 

 ning the oxygen as the loss by strong ignition he found as a mean of 

 9 closely agreeing experiments 9.44 oxygen for 100 DigOa, which very 

 nearly corresponds to the formula DioOj. However the atomic weight 

 of the didjanium in the oxide, used by Brauner, was 146.6 which indi- 

 cates that the oxide, regarded by Brauner as pure, really contained a 

 large percentage of samarium, the atomic weight of didymium being 

 142.3 and that of samarium 150. I found it therefore necessary to re- 

 peat Brauners experiments with the purest didymia I had. I heated the 

 basic nitrate in a platinum-basin till the red vapours ceased to be given 

 off. The residue was afterwards powdered, heated again, and lastly 

 heated in a current of oxygen to incipient redness. The material used 

 for experiments 1 and 2 were obtained in this manner. In all the other 

 cases I followed the description given by Mr. Brauner exactly. 

 My experiments are as follow: 



The loss on ignition was thus considerably less than in Brauners 

 experiments. Although I made 10 experiments I could not get more 

 than 5.02 in any single experiment. 



One may perhaps conclude that the reason for this difference is 

 that the material used by Brauner contained samarium, but it cannot 

 be so. In fact I am unable to suggest a satisfactory explanation. If 

 the loss on ignition be considered peroxide-oxygen my experiments tend 

 to the formula DiOs, which requires 4.82 parts of oxygen to 100 DijOg. 



Still I am very much inclined to doubt that the loss by ignition 

 is due to oxygen in chemical combination. I have tried to determine 



