Oxyddoride of Anthnony. 45 



By calculatiou. By experiment. 

 § CI = 6.639 = 11.49 = 11.32 



Sb = 8.064 ^ 

 ,5 f Sb = 36.290 } = 76.72 = 76.6 

 I i =. 6.75 = 11.79 = 12.08 



57.743 100 100 



Or two atoms of chloride are combined with 9 of oxide, the 

 formula being 2(3 CI + 2 Sb) + 9 Sb" The quantities of chlo- 

 rine and antimony found by experiment are, as was to be ex- 

 pected, a little less than is indicated by theory, causing the 

 amount of oxygen to appear something greater than it ought to be. 



Other compounds of chloride with oxide of antimony may pro- 

 bably exist. The composition of the powder of algaroth, whether 

 in the form of a dull grey sandy powder or of regular crystals 

 is very constant. In making the above analyses I employed the 

 crystals obtained at many different times, and the results agree 

 very closely. By supposing a portion of the acid to have been 

 washed away, it is easy to reconcile with the composition above 

 given all the results found by other chemists, with the exception of 

 that given by Dumas. If any confidence is to be placed on 

 that result, there must be at least one other analogous compound, 

 differing very much in composition from the common powder of 

 algaroth, the chemical constitution of which it is said to repre- 

 sent. 



There exist many other compounds of fluorides, chlorides, 

 and iodides with oxides, in which the ratio of the haloid com- 

 pound is to that of the oxide, so far as they have been examined, 

 as 1, 2, 3, 5, to ]. None of them, however, I believe, has yet 

 been obtained in a crystalliRe form, with the exception of the 

 oxychloride of antimony, and it is therefore not improbable that 

 considerable errors may hereafter be found in the atomic con- 

 stitution at present assigned to them. 



The difficulty of entirely decomposing the iodides of antimony 

 and arsenic even by long boiling with water, renders it exceedmg 

 probable that these iodides unite with their respective analogous 

 oxides in some definite proportion, which, of course, cannot be 

 investigated till they are obtained in a crystalline or some other 

 constant form. 



