482 Prof. Johnston on an Analogy in Atomic Constitution 



so long as we can represent the constitution of peroxide of 



iron by Fe+ Fe, a formula corresponding not only as a whole, 

 but also in each of its members, with that of titaniate of iron, 

 very little confidence can be placed in the support which this 

 case seems to lend to the method proposed. 



Still there is something very remarkable in the fact that 

 the earthy carbonates and the alkaline nitrates both crystallize 

 not merely in one but in tmoo forms wholly different, in each 

 of which forms they are isomorphous. The fact is so stri- 

 king, that we cannot lightly reject the notion, that a simple 

 relation must exist between the atomic constitution of the two 

 classes of compounds. 



In regard to the ultimate form of lime compared with that 

 of potash and soda, it is inferred, firsts that they are not iden- 

 tical because the analogous salts (the anhydrous sulphates for 

 example) crystallize in forms very different from each other* : 

 secondly, that the form of lime with an atom of water is pro- 

 bably identical with that of potash and soda, because with 

 this excess of water it appears to replace the latter in meso- 

 type, chabasie, and other minerals of the zeolite family. We 

 should expect therefore that where the same acid in the same 



proportion combines with K, Na, or Ca + H, the resulting com- 

 pounds should be isomorphous ; where it combines with Ca 

 only, the form of the salt of lime should be different. 



But though lime and potash combined with the same acid 

 A produce unlike forms, there is no reason why lime may 

 not combine with another acid B to produce a form identical 

 with that of potash, in union with A ; in other words, there is 

 no known law of crystalline combination in accordance with 

 which a form C can result from the union of two invariable 

 forms only, A and B. There may in all probability exist 

 other pairs or groups of forms D, E, differing to any extent 

 from A and B, yet so related that the result of their union 

 may be the form C. It need not surprise us therefore that 

 instances should occur, in which, though an isomorphic re- 

 lation be observed to prevail between two compound bodies, 



* The difference between the forms of the anhydrous sulphates of lime 

 and soda might be explained by supposing them dimorphous {Cristallogra- 

 phie von G. Rose, p. 159.), but if soda be isomorphous with potash, and 

 anhydrite differ in form from both alkaline sulphates, the difference could 

 only be explained by supposing all of them to be trimor[)hous ; a supposi- 

 tion not at present to be entertained. The three anhydrous sulphates 

 of potash, soda, and lime crystallize in right rhombic prisms, in which 

 M on M' subtends angles of 120° 30', 123° and 100° 8' respectively.— 

 (Brooke). 



