108 WRITINGS OF JOSEPH HENRY. [1855- 



each other, and under different conditions of the atmosphere 

 are exceedingly varied and beautiful. To explain these 

 figures in a general way let us suppose three separate atoms 

 to be within the sphere of mutual attraction and free to 

 move; they will approach until they come within the sphere 

 of repulsion, and will then evidently be found in the same 

 plane at the angular points of an equilateral triangle, since 

 each must be at the same distance from each of the other 

 two. If a fourth atom be suffered to approach in the same 

 manner it will also arrange itself at an equal distance 

 from each of the three others at the apex of a regular tri- 

 angular pyramid of equal and similar faces. The next 

 symmetrical arrangement which could take place would be 

 in case a fifth atom were added; and if this were situated 

 on the other side of the base of the pyramid a regular six- 

 sided figure would result. We see from these examples 

 that regular geometrical forms are the necessary effect of 

 the undisturbed grouping of the atoms, though it is impos- 

 sible to deduce all the facts from considerations as simple as 

 those we have given above. To adapt the hypothesis to the 

 facts of the case we are obliged to assume that crystalline 

 forms are not the result of the approximations of single atoms, 

 but of molecules of more or less complicated structure. 



Though the exact representation of the groupings of par- 

 ticles of different kinds of matter has exercised the ingenuity 

 of a number of investigators, the theory is still in a very 

 imperfect condition. It offers however a rich harvest for 

 scientific culture, and a number of interesting conclusions 

 have been deduced from the crystallographic study of bodies, 

 particularly by M. Gaudin. We are obliged to suppose that 

 the primary molecules which enter into crystals are them- 

 selves of a geometrical shape, due to the arrangement of the 

 ultimate atoms of which they are composed, and such forms 

 are called the primitive forms of the crystalline molecules. 

 These primitive molecules vary in form and size, as we shall 

 see hereafter, and they vary also in these respects, in some 

 cases of their combinations. If the two salts we mentioned in 

 the commencement of this division of our subject — namely, 



