138 ISOMORPHISM. 



and the protoxides of nickel, iron and cobalt, is well marked 

 in the salts which they form with a common acid, and is par- 

 ticularly observable in the double salts of these oxides, such 

 as the sulphate of magnesia and potash,, sulphate of zinc and 

 potash, sulphate of copper and potash, which have all six 

 atoms of water and a common form. The sulphates themselves 

 of these bases differ, most of them affecting seven atoms of 

 water of crystallization, while the sulphate of copper affects 

 five ; but those with the seven may likewise be crystallized in 

 favourable circumstances with five atoms of water, and then 

 all assume the form of the copper salt, thus exhibiting a second 

 isomorphism like the arseniate and phosphate of soda. 



The peroxides of the same class of metals with alumina and 

 the oxide of chromium, which consist of two atoms of metal 

 and three of oxygen, also afford an instructive example of iso- 

 morphism, particularly in their double salts. The sulphate 

 of the peroxide of iron with sulphate of potash and twenty- 

 four atoms of water, forms a double salt having the octohedral 

 form of sulphate of alumina and potash or common alum, the 

 same astringent taste, with other physical and chemical pro- 

 perties so similar, that the two salts can with difficulty be 

 distinguished from each other. The salt is called iron-alum, 

 and there are corresponding manganese and chrome alums, 

 neither of which contains alumina, but the deutoxide of man- 

 ganese and oxide of chromium in its place, with the propor- 

 tions of acid and water, which exist in common alum. In all 

 these salts another substitution may occur without change 

 of form; namely that of soda or oxide of ammonium for the 

 potash in the sulphate of potash, giving rise to the formation 

 of what are called soda-alum and ammonia alum. 



Certain facts have been supposed to militate against the 

 principles of isomorphism, which require consideration. 



1. It appears that the corresponding angles of crystals re- 

 puted isomorphous are not always exactly equal, but are some- 

 times found to differ one or two degrees, although the errors 

 of observation in good crystals rarely exceed 10' or 20' of a 

 degree. But it has been shown by Mitscherlich that a diffe- 

 rence may exist between the inclinations of two series of similar 

 faces in different specimens of the same salt, of -5 9' ; while it is 

 also known that the angles of a crystal alter sensibly in their 

 relative dimensions with a change of temperature (page 3). 



