1859.] as investigated hy Light, lOo 



wave surface. They are the resultant axes of polarisation before 

 alluded to ; they have no statical significance in the crystal. 



But this subject enters here into the domain of mathematical analysis. 

 Indeed, though experiment has effected much, and mathematical inter- 

 I)retation more, for the establishment of the laws that control the 

 difficult phenomena exhibited by oblique crystals, the subject is still 

 one in which much has to be done, especially in determining the 

 true axes of elasticity in these crystals. 



The last subject touched on, and that briefly, was the impossibility 

 of explaining the growth of a crystal by any architectural view, like 

 that of liafiy, i.e. by any view that supposes the crystal developed by 

 the addition, one by one, as it were, of molecules, endowed only with 

 forces acting at minute distances and depositing themselves by virtue 

 of these alone in th^ir position. The facts connected with the growth 

 of a crystal in its mother-liquor point to quite a different conclusion. 



The simultaneous production of the corresponding facettes, 

 however minute they may be, and however complicated the crystal form, 

 upon the opposite ends or parts of it, and in the precise positions 

 where symmetry requires them, needs for its explanation something akin 

 to an instinct in the molecules, if we are to suppose them so to deposit 

 themselves, as that their deposition is independent of influences extend- 

 ing at once over the whole crystal, and to the mother liquid investing it. 



Hemihedrism points to the same result.* 



So too how are the infinite numbers of tesselated crystalline frag- 

 ments that interpenetrate without any symmetrical orientation so many 

 crystals [Leydolt's Etching Experiments on Quartz, Apophyllite,&c.] 

 without interfering with their general form, to be explained ? 



Finally, the forms of crystals, say of fiuor spar, or of calc spar, 

 from the same mine, are similar, while those met with in a neighbour- 

 ing locality, where the conditions of deposit were different, are different 

 from these. The typical forms of barytes as found in Cumberland, 

 or in Auvergne, Scheranitz, &c. are different for each locality, and 

 so are often the hornblendes that occur in different rocks. Over large 

 extents of country a mineral {e.g. the augite of Southern Tyrol) will 

 present constantly the same combinations of crystalline form. Salt is 

 deposited in cubes from its simple aqueous solution, — in octahedra 

 when that solution contains uric acid ; and alum presents the form of 

 the cube when alumina is present in excess. 



In all these cases, therefore, — indeed, in every case — the growth of 

 a crystal is an inexplicable thing, so long as we endeavour to trace its 

 cause to powers residing in and confined to the molecules. A crystal, 

 like a plant, is developed in a medium, and as the plant owes the 

 special peculiarities of its individual form, notwithstanding the seem- 

 ingly perfect freedom of its growth, to special circumstances in the 



♦ Grailich connects hemihedrism {i.e. the development of only the alternate 

 planes required by the symmetry of the system of a crystal) with a difference in 

 P')s»tion of the centres of gravity ami centres of volume in the molecules. — {Private 

 Letter.) 



