4-12 llev. J. B. Kimnett on ihr 



expansion nor contraction, except such as results from the 

 receding from, or approach to, each other of the particles, in 

 right lines joining the centres of the adjacent particles. 



With respect to solids, the case is very different ; the utmost 

 expansion is very small, compared with that of the same bodies 

 in a liquid state. Some have supposed (Lavoisier's Chemistry) 

 that the particles of solids do not touch each other. Boscovich 

 imagined them to be separated to a distance from each other, 

 in a point of etjuilibrium between their own centripetal force 

 and the repulsion of caloric ; — nearer to the particle, he sup- 

 posed the force of repulsion to prevail; beyond it, that of at- 

 traction, which increases, according to his system, according 

 to some function of the distance directly, to a certain maxi- 

 mum : hence, if any force be applied, tending to separate the 

 particles, since their distance is somewhat increased, the cen- 

 tripetal force begins to produce sensible effects; when the 

 particles are removed to a distance beyond that at which the 

 force of attraction attains its maximum, they separate. Were 

 this the true state of the case, the particles of solids must have 

 the same freedom of motion which those of liquids possess ; or 

 in other words, there could be no solid in nature. It is also 

 evident from sect. 12, 13, of the first volume of Newton's 

 Principiay that whatever law of variation the centripetal force 

 obeys, the particles of solids nuist be in contact, otherwise the 

 observed phtenomena cannot be produced. 



These departments of science have received but little at- 

 tention from modern chemical philosophers, except so far as 

 the subject of crystallization is concerned ; and here, systems 

 are commonly received, which seem to be at variance with 

 established principles of physical science. For, spherical par- 

 ticles are so placed together, that if the centre of a particle be 

 joined with the centres of two adjacent ones, the lines form 

 anifles of 60° or 90° : since every variety of crystal cannot be 

 produced by such arrangements, some particles have been 

 supposed to be spherical ; others, ellipsoids, oblate and pro- 

 late, of various degrees of eccentricity. This being supposed, 

 a crystal cannot either expand or contract by change of tem- 

 pei'ature : for if it conti'act, the particles must be compressible; 

 if it expand, it is resolved into a liquid : this system cannot 

 account for the direction of the cleavages, nor explain why, on 

 heating the nucleus of a crystal (as a rhomb of carbonate of 

 lime), the acute angles are increased and the obtuse diminished. 

 Afain, since nearly all known crystals are compound bodies, 

 this system has to suppose a compound atom : /. e. a system of 

 several contiguous atoms, to assume the form ola regular el- 

 lipsoid, or some such figure^ in all, except a lew cases. The 



arijument 



