1818.] respecting the Specific Gravity of Crystals. 131 



to be the case, and that not as the deduction from any compli- 

 cated train of reasoning, or any loose analogy, but as the direct 

 result of experiment. As this maybe regarded as the^.experi- 

 mentum crucis, upon which the whole argument hinges, I think 

 it will be proper to quote it at full length, in order that I may not, 

 unintentionally, be guilty of any omission or misrepresentation. 

 " I selected," says Mr. Daniell, " a mass of green fluor spar, trans- 

 parent, and perfectly free from the adhesion of any foreign ingre- 

 dient. From this I extracted, by mechanical division, the 

 following solids ; a tetrahedron, a rhomboid, an octohedron, and 

 a cuneiform or lengthened octohedron. I then proceeded to take 

 their respective specific gravities with a veiy delicate balance, 

 making use of every precaution to avoid any source of error. 

 They were as follows : 



Cuneiform octohedron 3' 100 



Octohedron 3-037 



Tetrahedron 2-909 



Rhomboid 2-904 



" The result of this experiment was, therefore, perfectly satis- 

 factory ; the specific gravity of the octohedral arrangements 

 exceeding that of the tetrahedral in a very sensible degree." 



A second experiment, or rather series of experiments, which 

 Mr. Daniell performed, was conceived " to confirm this conclu- 

 sion in a still more unexceptionable manner." He took a cube 

 of colourless fluor spar, then cut off four of its corners, and 

 afterwards formed it into a regular octohedron ; he ascertained 

 its specific gravity in each of these states, and also that of three 

 of the solid angles that were cut off; the results of these six expe- 

 riments are arranged by the author in a tabular form, as follows : 



Specific gravity of the cube 3-180 



4 corners cut . . 3-242 

 octohedron .... 3'261 



1st corner 3-115 



2d ditto 3-111 



3d ditto 3-125 



Upon this experiment the author remarks : " Here we have 

 the very same solid of perfectly homogeneous composition, vary- 

 ing in its different parts in specific gravity according to the cal- 

 culations of theory, while nothing could be more unexpected 

 than the fact itself, till the calculations pointed it out as a test 

 whereby the hypothesis must stand or fall." Mr. Daniell con- 

 cludes by remarking, that " numerous other experiments all 

 agreed in the result, that fluor spar increases in specific gravity 

 according as, in its division, we approach to the perfect octohe- 

 dral arrangement, and recede from the tetrahedral." 



Having thus given you a full account of Mr. Daniell's experi- 

 ments, and of the deductions which he forms from them, 1 shall very 

 briefly state the remarks which have occurred to me on the sub- 

 ject. In the first place, it must be admitted that Mr. Daniell 



12 



