AND RESEARCHES OF JAMES BMITHBON. 127 



tallization is usually effected. But those who are familiar with chemical 

 operations, know the sort of agglutination which happens between the par- 

 ticles of subsided and very fine precipitates, occasioning them, on a second 

 diffusion through the fluid, to settle again much more quickly than before, 

 and which is certainly a crystallization, but under circumstances very un- 

 favorable to its perfect performance." 



The recent discovery of the reduction of wrought-iron 

 from a fibrous to a granular state by a mechanical percus- 

 sion, especially at a certain elevated temperature, is a case 

 strongly illustrative of the views of Smithson on this ab- 

 struse and difficult subject. 



In the same paper (on the calamines) he has attempted to 

 show a simple definite relation to exist between the constit- 

 uents of this material. 



In attestation of the value of these observations by Smith- 

 son, we may cite Gregory Watt's paper on the basalts pub- 

 lished in the following year, (1803 :) 



" It has been most justly remarked by Mr. Smithson, that solution, far 

 from being necessary to crystallization, effectually prevents its commence- 

 ment; for, while solution subsists, crystallization cannot take place. It 

 may remain a question, whether previous solution be essential as a prepara- 

 tory means of obtaining by subsequent evaporation, the small parts of 

 bodies disengaged so that they may unite to form regular crystals. If by 

 solution be only meant that simple action of heat or water which merely 

 counteracts the force of aggregation, and relieves the molecules from their 

 bond of union with each other, it certainly is a requisite; but if by solution 

 be meant that action of affinities by which not only the force of aggrega- 

 tion is overcome, but the combinations which constitute the molecules are 

 destroyed, it obviously is not only unnecessary, but prejudicial to the crys- 

 tallization ; as a new sot of molecules must be formed, by a new combina- 

 tion of the elementary particles, before the formation of regular bodies can 

 take place. The suspension of the molecules ready to crystallize may be 

 correctly said to be merely mechanical. Though the mechanical action of 

 trituration can never be expected to resolve even the most divisible body 

 into its molecules, because the fractures will be at least as frequently across 

 the natural joints as in their direction ; yet, even by this rude method, 

 some perfect molecules may be disengaged ; for we find that water, passing 

 over large surfaces of silicious sand, finds some molecules of silex in the 

 state proper for aggregation, and even for crystallization. Mechanical sus- 

 pension in a fluid medium of such density that the crystalline polarity may 

 be enabled to counteract the power of gravity, is with justice considered by 

 Mr. Smithson the only requisite for the formation of crystals. 



" The particles of bodies apparently solid must be capable of fome inter- 

 nal motion enabling them to arrange themselves according to polarity, 

 while they are still solid and fixed as far as they have reference to the ordi- 

 nary characters of fluidity." 



The mode of examining calamines, adopted by Smithson, 

 was to subject them to heat, in order to expel water and 

 carbonic acid, and then to dissolve the residue in sulphuric 

 acid, drying the white vitriol thus produced, and estimating 

 the weight of oxide by that of anhydrous sulphate. This 

 estimation of a metallic oxide in its state of a dry sulphate, 

 enables the chemist to avoid two or three operose and 



