SCIENCE AS A SYMBOL AND A LAW 15 



these theories are defined conjecturally as being due 

 to some kind of mechanical motion and motive force, 

 as when heat is defined as consisting in molecular mo- 

 tions, or the rigidity of solids in molecular attractions 

 and repulsions. 



The motions and forces involved in these theories 

 can no longer be ascribed to sensible matter, but either 

 hypothetical bodies, such as the luminiferous ether, or 

 hypothetical parts of real bodies, such as molecules, 

 atoms, ethereal vortices, or other imaginary elements 

 of matter must be created. And to them are assigned 

 properties and laws resembling as closely as may be 

 those of sensible bodies. In explaining new facts, as 

 they are discovered, the attributes of the hypothetical 

 matter are modified, or such new ones assumed as may 

 best fit the case. Such mechanical hypotheses, not be- 

 ing based on experimental evidence, are held to fulfill 

 their purpose when these conjectural attributes explain 

 in the simplest and most plausible way the largest 

 body of known phenomena and when they anticipate 

 phenomena afterwards observed. The importance and 

 weight of these hypotheses increase with the number of 

 phenomena whose laws they express. 



Certain hypothetical theories, such as the wave 

 theory of light, have been undoubtedly useful, since 

 they have reduced complicated actions to a few simple 

 laws. And also they tend to combine all branches of 

 physics into one system in which the axioms of 



