ON THE FLOW OF SOLID BODIES. 255 



has reached the same conclusions by purely mechanical theories, and 

 from this moment, therefore, they can be considered as belonging to 

 science in the most complete manner. 



Another kind of research has occupied us for a long time ; it is that 

 connected with the planing of metals. The shaving which the tool 

 carries away, and which up to that time had not even attracted the 

 attention of practitioners, offers many remarkable peculiarities. Not 

 only does it often twist itself spirally, which is the result of the unequal 

 shortenings of the different files of molecules of which it is composed, 

 but speaking generally, it can be calculated by how much it will be 

 shortened in a given case. It is usually reduced about a fifth of its 

 original length under ordinary circumstances ; and we can also con- 

 sequently tell by how much its transversal dimension will increase 

 during the same time, its density not having varied to any sensible 

 extent. This shaving, which is the "transformed of the prism" carried 

 off by the graver, increases in thickness in an inverse ratio to the 

 variations of its length. With regard to those complex convolutions, 

 in which the definite forms no longer depend on those of the moulds, of 

 the drawing-frames, or of the punches of which we have spoken above, 

 but merely on the exterior and interior forces which are brought into 

 play, we have not yet been able to characterize them with certainty, 

 except in a small number of cases. Here, however, is a model, which 

 shows how matter, driven back by a tool of a peculiar and symmetrical 

 shape, passes from a square section to a triangular section of much 

 greater dimensions, under the influence of pressures transmitted in 

 passing through geometrical mediums, traced according to a theoreti- 

 cal diagram and belonging to analytically defined surfaces. Many a 

 discovery will be made in molecular mechanics by following up these 

 phenomena, which have been so entirely overlooked up to the present 

 time, and which are so closely allied to the properties of matter. 



Smelting is also one of the means that can be employed in this 

 attractive study of molecular mechanic. A bar of iron is simply a 

 mass of agglutinated filaments placed in juxtaposition, which proceed 

 individually from a determined nucleus. In proportion as the number 

 of the filaments is increased, and the mass formed by their juxtaposition 

 is stretched out, they become more and more tenuous, and it is easy, 

 by special means of oxydation, to discover, in a manufactured piece of 



