175 



Criticism of an Experiment Used to Determine the Com- 

 bining Ratio of Magnesium and Oxygen. 



James H. Ransom. 

 In some of the modeni laboratory manuals for use in general chem- 

 istry work an experiment is described whereby a weighed amount of 

 magnesium powder is oxidi/.ed in a covered crucible until a constant 

 weight is oI)tained. The increase in Aveight has been assumed to be due 

 to oxygen, and thus the ratio of the two elements in the oxide easily 



calculated. 



My students have performed this experiment during the last two years 

 but have not been able to secure sufficiently concordant results to make 

 it appear to them as illustrating the law of constant composition. 



f-'ome observations are readily made in performing the experiment. 

 The product, except perhaps at the surface, is not white, as is mague- 

 sium oxide, but of a gray color, due evidently to a mixture of substances 

 of different colors. Also tlie crucibles at the end of the experiment are 

 coated within with a black substance which can not be removed even ou 

 scouring with sand: and the crucibles lose in Aveight. 



Examination of the product of l)urning shows that on treatment with 

 small amounts of water ammonia is evolved, thus indicating that mag- 

 nesium nitride is one of the substances present. As in this compound 

 the ratio of the elements is 1:388 Avhile in the oxide it is 1:007 it follows 

 that from this standpoint the increase in weight must be less than the 

 theory. Again, on treating the product with fairly concentrated hydro- 

 chloric acid a disagreealdy odorous gas is evolved which at times is spon- 

 taneously combustilde. It is. witliout doubt, hydrogen silicide from mag- 

 nesium silicide formed by the action of magnesium on the crucible ma- 

 terial. On treating with acid as above described there always remains 

 a black insoluble amorphous residue mixed with white particles which 

 under a hand-lens look like silica. The black mass when heated ou 

 platinum foil changes to a white powder whicli resembles silica. Appar- 

 ently the black portion is silicon. It is conceivable that a part of the 

 silicon after being formed, and during the heating, is oxidized by the air; 

 and as it unites with nearly twice as much oxygen as does the same 

 weight of magnesium, it might equalize the loss of the oxygen content 

 due to the causes already indicated. Thus can be explained the nearly 

 theoretical results so often obtained. But at l)est these results must re- 



