produced by a change in the solvent. 291 



various causes, such as the influence of mechanical mixtures, 

 the nature of the liquid in which the solution is made, and 

 the influence of substances which combine with the solution 

 at the instant of crystallization ; but to those cases in which 

 there is a complete change in the crystalline system, or a pas- 

 sage from one form to another which is entirely incompatible 

 with it. 



M. Beudant made an interesting experiment of this kind with 

 nitrate of potash, which commonly crystallizes in a right rhom- 

 boidal prism, and nitrate of soda, which crystallizes in the form 

 of an obtuse rhombohedron. When the two salts are dissolved 

 together, the nitrate of soda will crystallize first, if it exists 

 in sufficient quantity, and the nitrate of potash will afterwards 

 crystallize in prismatic needles. In a short time, however, 

 rhombohedral crystals of the nitrate of potash are deposited, 

 and though these were found to contain only small and vari- 

 able quantities of nitrate of soda, yet M. Beudant has observed 

 them perfectly free of nitrate of soda. 



On another occasion M. Beudant obtained a result the re- 

 verse of this. Having made an aqueous solution, containing 

 more nitrate of potash than nitrate of soda, there were first 

 formed prismatic crystals of nitrate of potash, then rhombohe- 

 dral crystals of nitrate of soda, upon which were found pris- 

 matic rhomboidal needles of the same nitrate of soda, con- 

 taining some traces of nitrate of potash. Hence, M. Beudant 

 concludes, that the change of form in the first instance was 

 owing to the presence of the nitrate of soda in the solution, 

 and, in the second, to the presence of nitrate of potash. 



Other examples of change of system in crystallized substan- 

 ces I have had occasion frequently to point out. In 1816, I 

 was led to examine crystals of nitrate of strontian crystallized 

 in prisms and also in regular octohedrons. But upon sub- 

 mitting these salts to M. Berzelius, he assured me that the 

 one was a hydrous and the other an anhydrous salt. In like 

 manner, I found single crystals of sulphate of potash in regu- 

 lar six-sided prisms, (not the compound bi-pyramidal crystals,) 

 and in rhomboidal prisms ; but upon transmitting them to M. 

 Berzelius for analysis, I learned that the former was a double 

 salt, containing one atom of sulphate of potash, and one atom 



