200 Dr. C. V. Burton on a Theory 



these quantities to be determined by the distribution of strain 

 in the medium ; it seems possible, indeed, that the greater 

 part of the effective mass of a strain-figure might reside in 

 that region where attraction follows the Newtonian law. 



10. Collision. — If we bring the centres of two strain-figures 

 close together, so that the spheres A intersect sufficiently far, 

 we shall be partly superposing two strain-distributions of such 

 a type that the energy increases with the strain, so that re- 

 pulsion may, perhaps, ensue. Of course this is only a very 

 rough attempt at explaining the effects of collision, and takes 

 no account of any deformation which might be produced in the 

 strain-figures on bringing their centres to such close quarters. 

 But even were such deformation produced, the results of a 

 collision might still be of a very simple character. So long 

 as the condition of § 4 is satisfied, so long, that is, as the 

 motions concerned are extremely slow compared with V (the 

 velocity with which gravitation is propagated) the distribution of 

 displacement in the medium will be at each instant the same as 

 if the two strain-figures were at rest. Hence, though the two 

 strain-figures may recoil from one another with altered velocities 

 they ivill not have acquired any vibratory motion from the 

 impact. This statement is equally true whether the strain- 

 figure is symmetrical or not, but in the case of symmetry 

 about a point, the properties of a strain-figure will be identical 

 with those of a Boscovich particle, exerting actions at a dis- 

 tance according to a law of force. 



11. The result just established has a bearing on the dyna- 

 mical theory of heat. We know that the number of degrees 

 of freedom of a molecule, and consequently also of an atom, 

 is probably finite. If an atom consisted of n strain-figures of 

 the most general kind, the total number of degrees of freedom 

 would be 6n ; but if each of the constituent strain-figures 

 w r ere symmetrical about a point the number w r ould be reduced 

 to 3n. Since the spectra of elementary vapours teach us that 

 the number of degrees of freedom, though finite, is usually 

 very large, we are led to infer that most atoms are formed by 

 the aggregation of a large number of strain-figures. 



12. Discrete Nature of Atoms. — Why do the atoms form a 

 discrete series, and why are all atoms of a given element 

 identical in physical and chemical properties? These are 

 questions which it would be difficult to answer, though it 

 does not seem quite so difficult to point out a direction 

 in which we may possibly look for an explanation. If an 

 atom consists of a number of points endowed with inertia 

 and with the power of exerting actions at a distance, it is 

 hard to see how any explanation can be given : nor is the 



