42 0» Numerical Proportions. 



blage of a determinate number of molecules in a determinate si- 

 tuation, containing among them a space incomparably greater 

 than the volume of molecules; and in order that this space may 

 have three dimenyions comparable between each other, a particle 

 must consist of at least four molecules. In order to express the 

 respective situation of the molecules in a particle, we must con- 

 ceive by the centres of gravity of these molecules to which we 

 may suppose them reduced, planes situated so as to^lcave on one 

 and the same side all the molecules which are beyond each 

 plane. Supposing that any molecule is not contained in the space 

 comprized between these planes, this space will be a polyhedron, 

 of which each molecule will occupy a summit, and it will be 

 sufficient to name this polyhedron, in order to express the re- 

 spective situation of the molecules of which a particle is com- 

 posed. I shall give to this polyhedron the name of representa- 

 tive form of the particle. 



Crystallized bodies being formed by the regular juxtaposition 

 of particles, mechanical division will therein indicate planes pa- 

 rallel to the faces of this polyhedron ; but it will be able to in- 

 •dicate others resulting from the various laws of decrement: be- 

 sides, there is nothing to hinder the latter from being frequently 

 more easily obtained than a part of the former ; and hence me- 

 chanical division may rather furnish conjectures, and conjectures 

 onlv, for the determination of the representative forms. There 

 is another way of ascertaining these forms : i. e. to determine 

 by the relation of the component parts of a body, the number of 

 molecules in each particle of this body. For this purpose I seC 

 out on the supposition that in the case where bodies pass to the 

 state of gas, their particles alone are separated and removed 

 from each other by the expansive force of caloric to distances 

 much greater than tliose which the forces of affinity and cohe- 

 sion have an appreciable action, so that those distances depend 

 only on the temperature and pressure which the gas supports, 

 and at equal pressures and equal temperatures, the particles of- 

 all the gases simple or compound, are placed at the same di- 

 stance from each other. The number of particles is on this 

 supposition in proportion to the volume of the gases*. What- 

 ever may be the theoretical reasons which seem to support it, it 

 can only be considered as an hypothesis ; but on comparing the 

 consequences which necessarily result with the phaenomena, or 

 the properties which we observe ; if it agrees with all the 

 known results of experience, if we deduce consequences which 

 are confirmed by ulterior experiments, it may acquire a de- 



* I Viave learned since the drawing up of my paper, that M. Avogrado 

 has made this last idea t!ie £;roundwork of an incjiiiry into the proportions 

 of the elements in chemical combinations. 



giee 



