Mr Pocklington, On the Kinetic Theory of Matter. 291 



and place a massless particle on it which is attracted towards the 

 axis of x with a force which is always normal to this axis, towards 

 it, and small. Then to a first approximation the motion is un- 

 affected (we assume that the initial kinetic energy is not small) 

 and a final state of rectilinear motion is tended to. Making a 

 closer approximation to find the subsequent motion, it is clear 

 that the tricycle describes a sinuous curve about the axis of x. 

 On account however of the tendency of the energy to pass into 

 energy of translation, this motion will ultimately become motion 

 of translation along the axis of x. The motion may be in either 

 sense. However the body be started, its motion will tend to one 

 of the two possible final states. 



We can treat the case of §§ 5, 6 similarly, placing the particle 

 at the centre of gravity of the body, and we then find that there 

 are only two possible final motions. 



12. We can of course explain Boyle's and Charles' laws for a 

 gas by any form of kinetic theory that postulates atoms moving 

 through the ether without producing much disturbance in it. 

 The difficulties met with are those connected with reversibility 

 and with the distribution of energy, especially the particular case 

 of the relation between the energy held in the ether as radiant 

 heat and that held as energy of translation of the molecules. 



An actual gas when disturbed from its normal state always 

 returns to it, hence the motion appears to be irreversible and 

 hence non-mechanical. This difficulty does not exist if we reject 

 Assumption B, for we have shown that reversibility is consistent 

 with a tendency to pass to a definite state of motion, and to regain 

 it if disturbed. Probably there is no real difficulty even if the 

 assumption is accepted, for we can easily suppose that in order to 

 obtain much increase of divergence from the normal state, the 

 initial circumstances must be very accurately adjusted, and that 

 even then the increase of divergence is only temporary, and is 

 succeeded by a decrease to the limit zero. 



The difficulty that each degree of freedom of the molecule, 

 including the vibrational degrees of freedom corresponding to the 

 emission or absorption spectra, should have an equal share of 

 kinetic energy on the average, probably cannot be overcome if 

 Assumption B is accepted. The more serious one that the ether 

 should hold a share of the energy infinitely large compared with 

 that held by the translational degrees of freedom of the molecules 

 certainly cannot. On rejecting the assumption, we have however 

 found cases where the whole of the energy passes into one degree 

 of freedom, and there is no reason for thinking that the passage 

 of nearly all the energy from the ether to the matter is irre- 

 concilable with a mechanical theory of ether and matter, or for 

 thinking that a mechanical theory of radiation necessarily leads 

 to an infinite value for the radiation of a black body. 



