METALLIC OXIDES. 243 



a. In order to determine the quantities in the composition Formed with 

 of this oxide, I digested ten grammes of antimony with nitric £ ,"ded°by ni-*" 

 acid until they were completely corroded. I then mixed the trie acid, 

 liquid with much water, and washed the precipitate with 



water uiml that fluid came off without being capable of red- 

 dening turnsole. The oxide thus obtained weighed, when 

 dry, 12065 grammes j and, in order to drive off all water, 

 I exposed it in a glass capsule to an heat not as high as ignition, 

 but it took fire on a sudden, and continued to bum like fungus, TnV.es fire by 

 (or tinder) at the same time subliming inn thick white smoke, l^'J^' *** 

 of which part was condensed on the sides of the glass. The white, 

 powder, by this means, became as white as snow,, and weighed 

 J 2*3 grammes. 



b. As this analytical method did not appear very good, I Another me- 

 mixed in a small glass retort ten grammes of murias hydrargy- |„f -^ ^ a^J 

 ricus (corrosive sublimate) in \ .-.>•■ 'der, with twenty grammes of corros. subli- 

 powdered antimony. The atm spttertc air of the retort hav- mate mh,th '°- 

 ing been expelled by hydrogen gas, and a small receiver rilled 



with hydrogen gas being also applied, I gently heated the mix- 

 ture until the murias stibio.sus (butter of antimony) came over, 

 and lastly 1 heated the body of the retort red hot, to distil over 

 the mercury amalgamated with that part of the antimony which 



had been added in excess. The quantity of 16,98 grammes and from the 



r , . . ^, " , quantity of 



of antimony remained in the retort. Consequently ten t \, e j att * er de- 

 grammes of murias hydrargyricus had been decomposed by composed by 

 302 grammes of antimony. But 100 parts of this salt contain ^^tv f 

 5"75 of oxigen combinable with other metals. ICO parts ofamim,tae 

 antimony had, therefore, been combined with thirteen parts ^'"{^"^je 

 of' oxigen. were deduced, 



I repeated this experiment several times without having Vlz * l yQ^p?rts 

 ' l to an tiro, ana 



ever obtained results perfectly equal, but varying, as for example, about 19§ oxi- 



3 g,35 or 19.68 parts of oxigen for 100 of antimony The S eu - 

 causes of error in this experiment may be several V >r in- 

 stance, it is possible that the mercury may be so adherent to the 

 antimony as not to be separated but at a teinpeui ;> Which 

 would also carry over a little of the latter ; and ir is also pos- 

 sible that a small quantity of mercurial muriate may arise be- 

 fore decomposition along with the vapors or the muriate of 

 antimony. It is, therefore, probable, that thest experiments 

 may have given the quantity of oxigen rather too ^reat. 



In 



