\ 



The Evolution of the Sciences 



at a watch whose case they must not open, who 

 wonder how the watch manages to go. All that 

 we can require of a hypothesis is that it should 

 connect the known facts into a coherent system 

 consistent with the remainder of our knowledge. 

 It is impossible for us to go further. 



Consequently the question is not whether the 

 ether really exists or not, but what qualities must 

 be attributed to it in order to explain known 

 facts. Were we to imagine the ether as a gas 

 more tenuous even than hydrogen, we should be 

 on the wrong course and should explain nothing. 

 Assuredly ether must have a very low density. 

 According to the calculations of M. Brillouin one 

 gramme of ether should occupy a volume of, 

 roughly, one cubic kilometre. If we consider 

 that a like volume of air weighs 1,300,000 metric 

 tons, the ether will impress us as being practically 

 imponderable, but it is not a fluid. We may say, 

 however extraordinary it may seem, that it has 

 rather the properties of a solid. Lord Kelvin, 

 the admirable physicist, whose eyes saw farther 

 than the eyes of other men, compared it to a jelly 

 or to pitch. Like a jelly it quivers at the 

 slightest shock, and vibrations are propa- 



98 



