CHEMISTRY 



351 



it was ignited to oxide. As was to be expected (from the work of 

 Richards), this oxide was found to contain a fairh^ large, but surpris- 

 ingly constant, proportion of occluded gases, the proportion of which 

 was carefully determined in each sample of oxide analyzed. It was 

 further shown experimentally that nickelous oxide, at the temperature 

 at which it was ignited, about 1000°, possesses an inappreciable oxygen 

 pressure, and that oxide which has been ignited and cooled in an oxygen 

 atmosphere of a few millimeters pressure does not gain in weight upon 

 ignition and cooling in air, or even oxygen at atmospheric pressure. 

 Hence the treatment of the oxide preliminary to weighing was simply 

 prolonged ignition in air at about 1000°. After being weighed it 

 was reduced in a current of electrolj^tic hydrogen in a quartz tube at 

 about 1000°. No difficulty was experienced in securing constancy in 

 weight of the metal, which was assumed to indicate complete reduction. 



Table A. 



Terrestrial. 



Meteoric. 



Corrected 



weight of 



NiO iu 



vacuum. 



Corrected 



weight of 



NiO in 



vacuum. 



Weight of 



Ni in 

 vacuum. 



gm. 

 6.41281 

 6.10952 

 6.91023 

 4.60077 



5.57122 I 4.37792 ! 58.700 

 5.59217 I 4.39445 j 58.704 



Average | 58.703 



Average of all determinations . . . 



gm. 

 6 . 65624 

 4.45609 

 4.64795 



gm. 

 5.23021 

 3.50196 

 3.65212 



Average . 



Atomic 

 weight of 



Ni. 



58.683 



58.724 

 58.676 



58.694 

 58.700 



There seems to be no evidence of dissimilarity in the two specimens 

 of material, an outcome in accord with earlier experience of the same 

 kind. The results agree, as well as could be expected, with the one 

 obtained by Richards and Cushman, 58.68 (Ag= 107.880). 



Mr. Merritt R. Grose continued the analysis of cadmium bromide, 

 begun by Dr. Hartmann (see Year Book No. 13), by the electrolytic 

 deposition of the cadmium upon a mercury cathode. The glass cell 

 devised by Baxter and Hartmann for the analysis of cadmium chloride 

 was equally useful here. It was found advisable, however, to bubble 

 a slow current of hydrogen through the electrolyte in order to hasten 

 the removal of the bromine liberated at the anode. As with the 

 chloride, complete deposition of the cadmium was never secured, but 

 the residual electrolyte was carefully evaporated with sulphuric acid 

 and the very small amount of undeposited material was weighed as 

 sulphate. In some experiments the cadmium bromide was all con- 

 verted to sulphate before electrolysis. This modification had no per- 

 ceptible effect upon the results. 



