140 ISOMORPHISM. 



stitution, and contain a like number of atoms of their consti- 

 tuents. But it must be admitted that the most perfect coin- 

 cidence in form, or true isomorphism is likewise observed 

 between certain bodies which are quite different in composition. 

 Thus bisulphate of potash is dimorphous, and crystallizes in one 

 of the two forms of sulphur (Mitscherlich). Nitrate of potash in 

 common nitre has the form of arragonite, and occurs also., 

 there is reason to believe, in microscopic crystals in the form 

 of calc-spar. Nitrate of soda, again, has the form of calc-spar. 

 Hypermanganate of barytes and the anhydrous sulphate of soda 

 likewise crystallize in one form. Between the first pair, sul- 

 phur and bisulphate of potash, the absence of all analogy in 

 composition is sufficiently obvious, notwithstanding their 

 isomorphism. Between nitrate of potash and carbonate of 

 lime, and between hypermanganate of barytes and sulphate 

 of soda, there is no similarity of composition, on the ordinary 

 view which is taken of the constitution of these salts, but both 

 of these pairs have been assimilated, in speculative views of 

 their constitution proposed by Mr. Johnston* in regard to the 

 first pair, and by Dr. Clarkf in regard to the second, which 

 merit consideration, although the hypotheses cannot possibly 

 be both correct, as they are based upon incompatible data. 

 Besides these examples of identity of crystalline form without 

 any well established relation in composition, many others might 

 be quoted, if occurrence in the simple forms of the cube and 

 regular octohedron should be allowed to constitute isomorphism. 

 For example : carbon, sea-salt, arsenious acid and alum, all 

 occur in octohedroiis, although they are no way related in 

 composition. But these simple forms are so common, that 

 they can be held as affording no proof of isomorphism, unless 

 in cases where it is to be expected from admitted similarity 

 of composition, as between the different alums, or between 

 chrome iron and octohedral iron ore. 



But notwithstanding the occurrence of such apparently for- 

 tuitous coincidences in form, isomorphism must still be con- 

 sidered as the surest criterion of similarity of composition 

 which we possess. Truly isomorphous bodies generally cor- 

 respond in a variety of other properties besides external form. 

 Arsenic arid phosphorus resemble each other remarkably in 



* Philosophical Magazine, third series, vol. 12, page 480. 

 t Records of General Science, vol. iv, page K>. 



