like crystallineforms indicate like Chemical Formula:. 411 



which might be represented by the same general formula. 

 It was natural, therefore, to reason in an inverse order, and 

 to ask, if the same formula so generally follows or attends 

 upon the same form, may not all compounds of which the 

 form is alike be represented by like formulae ; — when put in a 

 rational form, should not the equivalents deduced from a cor- 

 rect analysis give like formulae for minerals alike in form ? 

 In this idea was presented a new and beautiful test of the 

 accuracy of mineral analyses. Guided by it, Abieh, under the 

 direction of Mitscherlich, examined the family of octohedral 

 minerals, of which magnetic iron is the type, and found that 

 spinella, gahnite, chrome, iron, &c. may all be represented 



by the same general formula R R. Other cases of a similar 



kind afterwards occurred, so that it began to appear, and to 

 be thought by many, that like forms always indicate like 

 formulae. 



7. But then came forward one after another the anomalies 

 or exceptions to this generalization which are grouped to- 

 gether in the above table, and in which the attendance of like 

 formulae on like forms cannot be distinguished. 



By these exceptions, however, the doctrine of isomorphism is 

 not touched, or the evidence in its favour in any vvay impaired. 

 They establish, or seem to do so, merely the negative posi- 

 tion, that like forms do not always in compound substances 

 indicate like formulae; an important and independent truth, 

 which points to some more general law we have still to 

 seek for, but which is perfectly consistent with the fundamental 

 and leading principles of the isomorphic doctrine. 



In the two papers published in former numbers of this 

 Journal*, in reference to the second and fourth of the groups 

 in the table, I was preparing the way for the consideration of 

 the general question which has been the subject of the above 

 remarks. The present paper I have delayed till after the 

 publication of the 6th volume of the Reports of the British 

 Association, that I might not anticipate my own report, of 

 which the above table forms a part. Meanwhile my friend 

 Mr. Richard Phillips, shocked at the changes which one of 

 the cases I had considered seemed to indicate, hastened to 

 publish his " Observations on Isomorphismf," some of the 

 remarks in which, I am satisfied, he will now see to have been 

 unnecessary. In reply to the fourth paragraph of these ob- 

 servations, the only one which calls for remark, I would ob- 

 serve, that judging from my own feelings, I am certain that 



• Vol. xii. pp. 324 and 480. f Ibid., p. 407- 



