OF ZINC AND ANTIMONY. 365 



law. Again, the crystals of SbZiig take up an excess of zinc, but do not take up an ex- 

 cess of antimony, while those of SbZug crystallize with an excess of either, facts which 

 are as inconsistent with the idea of mechanical impurity as the last. Finally, the form 

 of the curve of ShZn^ of itself alone proves that the excess of antimony in the crystals is 

 not in the condition of mechanical impurity, for in that case the variation of composi- 

 tion would not be influenced, as the curve shows that it is, by the chemical force. 



A theory that the variation in composition resulted from the mixture of two or 

 more isomorphous compounds, would be even less tenable than the one just discussed, 

 for, in the first place, it would be necessary to assume the existence of two other com- 

 pounds of zinc and antimony isomorphous with ShZn^, and of one other, if not of more, 

 isomorphous with ShZn^. Not only would such an assumption be contrary to all the 

 analogies of chemistry, and therefore require strong evidence to sustain it; but, in the 

 second place, it can almost be demonstrated that no such compounds exist. The crys- 

 tals having the calculated composition of either SbZna or SbZuj are marked, as has 

 been shown, by striking peculiarities, and, with one possible exception, similar peculi- 

 arities were not observed throughout the whole series of crystals which have been ex- 

 amined. The crystals containing 50 per cent of zinc and of the composition of SbZn4, 

 were found to have a slightly smaller Sp. Gr. than those just above or just below them ; 

 but the difiierence is so small that it may be accidental, and as the crystals exhibited 

 none of the other peculiarities which characterize crystals having the calculated com- 

 position of SbZuj or SbZnj, I could not attach sufficient weight to the one circum- 

 stance to feel authorized in admitting a third compound of zinc and antimony. Admit- 

 ting, however, the existence of SbZn^, yet, as exactly the same angle has been observed 

 on crystals containing 55 per cent as on those containing 43 per cent of zinc, it would 

 be necessary, in order to explain the variation in composition by the principle of iso- 

 morphous mixtures, to assume the existence of still a third compound isomorphous 

 with SbZuj, and containing more zinc than SbZuj, which would increase greatly the im- 

 probability of the theory in question. Again, the only probable compound of zinc and 

 antimony containing less zinc than SbZug would be SbZn ; and it will be remembered 

 that the crystals of SbZuo which contained the largest excess of antimony corresponded 

 very nearly to this compound. In like manner the crystals of SbZn^ Avhich contained 

 the largest excess of zinc corresponded very nearly to SbZuj. If, then, the excess of 

 antimony or zinc in the crystals of SbZuo arises from a mixture of isomorphous com- 

 pounds, it must be that SbZuj, SbZua, and SbZn are isomorphous. That the first two 

 are not isomorphous may be seen by turning back to the description of their crys- 

 talline form, and that there is no crystalline compound SbZn is sufficiently proved by 



