346 Mr. J. Barton on the Physical Causes 



supply this deficiency, it may be useful to revert to an expla- 

 nation above given of an observation of Mitscherlich^ on the 

 unequal expansion of crystallized bodies in different directions* 

 It was suggested that the particles of heat do not accumulate 

 equally round every part of a particle of solid matter if its 

 axes are of unequal length. Now, this inequality of distribu- 

 tion is such, by the nature of the case, as to counteract the 

 tendency of the particles to place themselves in the position 

 of greatest attraction. In other words, the compound force 

 exerted' by the central solid nucleus, with its surrounding at- 

 mosphere of heat, will approach more and more nearly to that 

 of a sphere as the heat increases ; and by consequence the 

 figure of the particles exercises a less and less influence on 

 their mutual position. When that influence. is completely 

 neutralized, the body is a fluid. 



8. As the mutual attraction of the particles is not however 

 destroyed in fluids, these particles still arrange themselves in 

 the position of greatest attraction.. But this position is no 

 longer influenced by their figure; it is now such precisely as 

 would be assumed by them if spherical, that is to say, the 

 position in which the distance of their centres is a minimum. 

 Now this is the position in which the sum of the intensities 

 between them is also a minimum ; therefore the whole bulk 

 of the fluid just before congelation is a minimum. If we sup- 

 pose other forces to come into operation, those, for instance, 

 resulting from the figure of the particles, the dimensions of 

 the body will consequently be enlarged ; and thus it appears 

 why so many substances are found to expand in the act of 

 congelation. 



9. Further, as the position of greatest attraction amongst 

 the particles of a solid is that in which their salient angles are 

 presented towards each other, it follows that at the moment 

 of solidification, when this arrangement is established, the 

 orbits of a certain portion of the accompanying particles of 

 heat will be changed from the ellipsoidal to the hypei^boloidal 

 form ; that is to say, a certain quantity of latent heat is con- 

 verted into sensible heat. 



10. Those solid bodies which are composed of the largest, 

 or rather the heaviest particles, are at once the best con- 

 ductors and the w orst radiators of heat ; for the velocity in- 

 creases with the mass. Hence the metals are the best con- 

 ductors of heat, and as far as the experiments hitherto made 

 enable us to judge, their conducting power seems to follow 

 the same order with the magnitude of their component atoms, 

 the differences not being greater than may be reasonably sup- 

 posed to arise from errors of observation. For a similar rea- 



