364 ON TWO NEW CRYSTALLINE COMPOUNDS 



soon form a coating impregnable to the strongest acid. As has been shown in the 

 previous part of this memoir, when the antimony and zinc used are free from arsenic, 

 the hydrogen obtained by this process is chemically pure. It is, consequently, com- 

 pletely destitute of odor. 



General Conclusions. 



I stated at the commencement of this memoir, that I expected to be able to prove, 

 first, that zinc and antimony form with each other two, and probably only two, definite 

 compounds ; second, that these compounds are capable of a very large variation in com- 

 position without any change in the crystalline form ; and, lastly, that this variation can 

 only be explained by admitting an actual perturbation in the law of definite proportions 

 produced by the influence of mass. That zinc and antimony form with each other two 

 compounds, and that these compounds are capable of a very large variation in composi- 

 tion without any change in the crystalline form, have been shown to be facts. The cause 

 of this variation can only be inferred. Before stating the conclusions to which, as I 

 think, the facts now established directly point, it will be well to consider the only two 

 admitted principles of chemical science which could possibly be brought forward to ex- 

 plain similar variations. They are, first, that of impurities in crystals ; second, that of 

 isomorphous mixtures. It will not be difficult to show that the variations in com- 

 position of SbZuj and SbZug cannot be explained by either of these principles. 



It is a well-known fact, that crystals frequently take up impurities, which are either 

 dissolved or mechanically suspended in the menstruum in which they form, and it might 

 be supposed, at first sight, that the excess of zinc or antimony in SbZug or SbZug bore the 

 same relation to their crystals that the sand does to the rhombohedrons of Calcite from 

 Fontainebleau, or oxide of iron and Chlorite to crystals of Quartz ; but, in the first place, 

 in all cases where a considerable amount of impurity is present, the crystals are either 

 imperfect, or else the angle is considerably changed, at times even as much as two or 

 three degrees ; and, secondly, as such impurities are merely mechanical, the amount in 

 the crystals would in all probability be proportional to the amount present in the men- 

 struum at the time of their formation. Now, in the crystals of SbZuj from the alloy of 60 

 per cent of zinc, there is present an excess of zinc amounting to 15 per cent, and never- 

 theless the crystals are perfect, and their angles identical with those of the crystals 

 obtained from the alloy of 43 per cent. In the crystals of SbZug the excess of zinc is, 

 to a certain limit, directly proportional to the excess in the alloy ; but in those of SbZug 

 the excess of antimony is far from obeying this rule, and were the excess in both cases 

 a mechanical mixture, the variation in both cases would undoubtedly follow the same 



