294 Formation of Compound or Twin Crystals. ' 



There may be also adduced a satisfactory solution of the occur- 

 rences which are included under Dimorphism. M. Voltz has at- 

 tempted to deduce by mathematical considerations one of the forms 

 of a Dimorphous body from the other. His process is as follows : 



Let a, 6, c represent the force of attraction in the axes of a Rec- 



foll 



a, b, c 

 a, c, b 



we have 2a, b+c, b+c, three sums, of which 

 two are equal. They therefore represent the attraction in the axes 

 of a Right Square Prism,. the other form of the dimorphous body of 

 which the above Rectangular prism was one. By adding these 

 quantities again in the following manner : 



a, b, c 

 c, a, b 

 b y c, a 



i 



we obtain three equal sums a + b + c which therefore are the 

 axes of a cube, another form of the same dimorphous body. Thus 

 the form of Rutile has been determined by him from that of Ana- 

 tase, and the proportional dimensions of the crystals of Carbonate of 

 Lime, from those of Arragonite, in which he has approximated very 

 nearly to the truth. The process however appears to be quite too 

 mechanical, and probably if the deductions should prove correct, it 

 will be found that they depend on a different cause from that which 

 appears in the author's writings on this subject. > He says in ac- 

 counting for it, " the different polar forces combine with each other 

 as in a Chemical Compound." 



He even proceeds so far as to suppose that a cube results in all 

 instances from a combination of three inequiaxal solids. This 

 seems to be so unnecessary as well as improbable a supposition, if 

 permeation of particles is admitted, that I somewhat distrust wheth- 

 er my own opinions on this subject, and the discovery of the similar- 

 ity in many parts of his theory to my own previously formed views, 

 may not have led me to suppose incorrectly that his theory was 

 founded on this hypothesis. Still it is difficult to conceive, how the 

 remaining parts of the exposition of it are consistent with any other 

 hypothesis. 



