130 Remarks upon Mr. Darnell's Experiments [Feb. 



having had the good fortune to reduce to a certainty, and to 

 establish as an undoubted matter of fact, what had before been 

 advanced only as a plausible hypothesis. With respect to the 

 truth of the hypothesis, I was very much disposed to coincide 

 with Mr. Daniell, even before I read his experiments ; and with 

 respect to the experiments themselves, I consider them in their 

 consequences as, perhaps, even of more importance than the 

 ingenious author is disposed to attach to them. If you will, 

 therefore, allow me to trespass so far on your pages, I shall first 

 abstract an account of the experiments from Mr. Darnell's paper, 

 and then offer some brief remarks upon them. 



It is well known that there are certain crystals which, by dis- 

 section, are divisible either into the tetrahedral or octohedral 

 form ; both these figures are bounded by triangular faces, the 

 tetrahedron being contained by four, and the octohedron by 

 eight. Now if we assume the bases of these triangles to be com- 

 posed of the same number of particles, as the one solid is 

 bounded by four, and the other by eight of these triangles, the 

 author concludes " that the whole superficies of the latter is 

 exactly double that of the former." He then goes on to state, 

 " that solids so constructed must differ in their specific gravities, 

 unless the number of elementary particles in the octohedron be 

 exactly double the number in the tetrahedron ; that is to say, 

 unless the number of atoms in a given space be equal in both 

 arrangements. " It is, however, easily perceived, that if we pile 

 up a set of balls into the forms of a tetrahedron and an octohe- 

 dron, the number of balls in the former will not be half that in 

 the latter ; if, for example, we assume four balls as composing 

 the base of each triangle, the tetrahedron will contain only 20, 

 and the octohedron 44 ; the latter, therefore, contains "more 

 than double the number of particles under a double surface.'" 

 The conclusion which Mr. Daniell draws from this fact, and in 

 short that which forms the basis of his experiments, is, that 

 " the specific gravity of the latter solid (the octohedron), must 

 be greater than the specific gravity of the former (the tetrahe- 

 dron)." Now, in fluor spar, we have a method by which the 

 hypothesis can be at once put to the decisive test of experiment ; 

 for as we can at pleasure reduce this body to the form either of 

 a tetrahedron or an octohedron, we have it in our power to form 

 two crystals, which should differ in their specific gravity in the 

 same proportion as they differ in the number of their component 

 particles. The author states the question with perfect correct- 

 ness. " Is the specific gravity of a mass of fluor, split into the 

 form of an octohedron, greater than the specific gravity of the 

 same mass, split into the form of a tetrahedron ? " Or, gene- 

 ralizing the question, as I apprehend we may do, in strict 

 conformity with the author's principle, does the shape of a body 

 affect its specific gravity i Singular, and indeed startling as such 

 an assertion may at first view appeal', we find that this is stated 



