202 THE LIMITATIONS OF SCIENCE 



least conversant with their history knows that their 

 number is great, and that they are not proposed con- 

 secutively. As for their character, he evidently means 

 that they are dangerous if we forget they are mere 

 conveniences, fictions, not to be taken too seriously. It 

 is perfectly safe to say that there is scarcely a text- 

 book or a treatise in any science which does not state 

 explicitly that the sort of things classed by Poincare 

 as neutral hypotheses, are realities and not conjectures. 

 Before showing how Poincare and Sir Oliver Lodge 

 confirm this opinion, let me quote two statements of 

 Sir J. J. Thomson: 



" The ether is not a fantastic creation of the specu- 

 lative philosopher ; it is as essential to us as the air we 

 breathe. . . . The study of this all-pervading sub- 

 stance is perhaps the most fascinating and important 

 duty of the physicist." 



" The possession of a charge by the ions increases 

 so much the ease with which they can be traced and 

 their properties studied that, as the reader will see, we 

 know far more about the ion than we do about the un- 

 charged molecule." 



Evidently convenience of computation and an aid to 

 our understanding by concrete images are not at all in 

 the minds of even eminent physicists when they indulge 

 in neutral hypotheses of ethers and atoms. 



Now let us return to Poincare, and see how he heeds 

 his own warning. But first notice how clear and how 



