128 Remarks on Mr. Daniell's Theory. [Feb. 



discovering the longitude, an experimentum crucis came across 

 rny mind. 



I procured two cubes of lead precisely equal in weight, and of 

 the same specific gravity 5 every angle was determined "by the 

 reflecting goniometer to be 90® within less than a second ; and 

 having placed them on my table, I divided them both into two 

 equal parts, the one, by a vertical section through the diagonal, 

 the other, by a vertical section parallel to one of the sides. The 

 former cube was thus converted into two triangular prisms, the 

 latter into two quadrangular prisms. 



Now, without going into the minutiae of calculation, it is evi- 

 dent that the diagonal of a square being to its side as the \f 2 : 1, 

 I obtained a much larger proportion of new surface by the diago- 

 nal, than by the parallel division ; for the original surface of the 

 cubes being in no wise altered, you will easily see that the surface 

 of the triangular prism is to that of the quadrangular :: 4*414 : 4. 

 Here it is obvious, said I, repeating the words of Mr. Daniell, 

 (which I have by heart) that solids, so constructed, must differ in 

 their specific gravities, unless the number of particles in the one 

 exceed the number of particles in the other, in the same propor- 

 tion that the surface of the former exceeds that of the latter. 



I confess to you, Sir, that I proceeded to ascertain the weights 

 of these demi-cubes with some palpitation of heart, fearing that 

 their new surfaces might have some influence on their weight in 

 air, which would have exceedingly disconcerted my experiment. 

 Luckily, however, this was spared me ; they were all precisely of 

 the same weight. 



" I then proceeded to take their respective specific gravities," 

 " making use of every precaution to avoid any source of error," 

 when, to my utter astonishment and surprise, there was not the 

 slightest difference whatever. My scales had been made by the 

 late Mr. Coventry, with conical beams, ruby points, and agate 

 planes, and turned easily with the _._>_ 5 .th °* tne we *g nt 5 y^» 

 after innumerable trials, I have never been able to detect a varia- 

 tion as tar as the 10th decimal figure, beyond which I have not 

 thought it necessary to proceed. 



My faith is not. staggered by this untoward result ; but I ac- 

 knowledge myself puzzled, and fly to you for information. I 

 humbly request that either you, Mr. Editor, or some of your 

 readers, or Mr. Daniell, if this should meet his eye, will give 

 ine some explanation of these provoking facts. 



As this is a question of surface, can an incipient oxidation of 

 the surfaces have given rise to the error? If so, the experiment 

 :night be made in a more refined manner with platina cubes. I 

 have turned to the books on the subject; but neither Galileo, nor 

 Boyle, nor Hooke, nor Newton, nor Cotes, nor Huygens, nor 

 the Bernouillis, nor any writer, ancient or modern, that I hava, 

 consulted, ever once mention* the word surface in reference 



