248 KINETIC THEORIES OF GRAVITATION. 



concomitant knowledge, but if it be a reality, must forever remain to 

 us incomprehensible." How the Bchool-boy's personal experience of the 



strain excited in drawing by a cord his winter's sled can be resolved 

 into a sensation of " pressure" does not seem easy of discovery. 



Assuming then an order of a'therial waves having a much larger 

 range than those of light, Professor Challis endeavors to deduce the 

 several laws, of action proportional to the number of atoms acted upon, 

 of the inverse square of the distance of action, and of simultaneous 

 action in different directions. lie infers that if such waves encountered 

 a slight retardation of propagation in passing through the earth, they 

 would be refracted, so to speak, by the form of the large inequalities of 

 the terrestrial surface, producing the observed deflection of the plumb- 

 line. 



He also supposes that a small function involving r 4 must be added 

 to the usual formula ^ increasing sensibly the action of the sun near 

 its surface and diminishing its action notably through interstellar dis- 

 tances, lie thinks that this alone will explain why the sidereal system 

 does not collapse toward its common center of gravity. "According 

 to the theory of gravity I have proposed, although the ordinary law 

 may be exact through the solar system and far beyond, there must be 

 distances at which the condition that the excursions of the vibrating 

 particles of the aether are large compared to the dimensions of the 

 atoms ceases to be fulfilled. In that case the attraction changes to 

 repulsion."* 



lieuewing the discussion of "A Theory of Molecular Forces," the fol- 

 lowing year, Professor Challis contends: " It is a matter of demonstra- 

 tion that a theory of molecular forces cannot be constructed on the 

 hypothesis that the forces vary according to some law of the distance 

 from individual material particles, unless the law be such that the force 

 changes sign with the distance so as to become attractive alter being- 

 repulsive. But if force be a virtue resident in the particle, it must at its 

 origin be either attractive or repulsive, and it seems impossible to con- 

 ceive how by emauation to a distance it can change its quality. This 

 difficulty, as will be shown, is not encountered in a theory of molecular 

 forces which deduces their laws from the dynamical action of an elastic 

 medium." While it is probably no more difficult to conceive an innate 

 force or virtue which " at its origin " shall have a law of radial intensity 

 whose value passes through zero, than to conceive any other mathemat- 

 ical law of increment or decrement, there is certainly no necessity for 

 assuming such a law. If we should suppose the attraction by inverse 

 squares to be absolute, with the superposition of a repulsion of much 

 higher inverse power, and of far greater intensity, it is evident that the 

 two curves of force would cross each other, and that at the intersection 

 tin- resultant would involve a " change of sign." However difficult it 

 may be to r ealize such a conception, the actual superposition of oppos- 

 * Phil. May., ld5 ( J, vol. xviii, p. 401. 



